JP2016028346A - Artificial intelligence device for expanding processing capacity in self-organizing manner - Google Patents

Artificial intelligence device for expanding processing capacity in self-organizing manner Download PDF

Info

Publication number
JP2016028346A
JP2016028346A JP2015205904A JP2015205904A JP2016028346A JP 2016028346 A JP2016028346 A JP 2016028346A JP 2015205904 A JP2015205904 A JP 2015205904A JP 2015205904 A JP2015205904 A JP 2015205904A JP 2016028346 A JP2016028346 A JP 2016028346A
Authority
JP
Japan
Prior art keywords
pattern
information
sentence
patterns
relationship
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP2015205904A
Other languages
Japanese (ja)
Inventor
洋彰 宮崎
Hiroaki Miyazaki
洋彰 宮崎
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to JP2015205904A priority Critical patent/JP2016028346A/en
Publication of JP2016028346A publication Critical patent/JP2016028346A/en
Pending legal-status Critical Current

Links

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

PROBLEM TO BE SOLVED: To provide artificial intelligence that can perform transitions of logics and transitions other than those of the logics such as transitions of human thought patterns, can allow humans to understand processing courses, and is not ruined.SOLUTION: An artificial intelligence device performs: processing for analyzing and recording information and the structure of the information by using human's instruction and learning; building of relations between pieces of the information as an information system by connection relations between patterns and processing between the patterns; learning of a method for processing input information, such as a method for analyzing the relations between the pieces of information and the pieces of information, a method for building the information system, and a method for solving problems; organizing of the input information into sentence structures by the analysis of it; autonomous recording of information determined to be useful to build the information as the information system; and autonomous processing (recording of the information, update and improvement of the information system, performance of instruction contents, and responses to questions) in accordance with results evaluated by verification with the information system.SELECTED DRAWING: Figure 11

Description

この発明は逐次、入力した情報から自律的に知識体系を構築するとともに、構築した知識体系を活用して状況の分析および問題の識別を実施し、会話、翻訳および問題を解決するための情報検索、処理を実施するプログラムの起動、問題解決に必要な情報の生成と知識体系への組み込みを実施し、自律的に問題を処理するとともに更なる知識の向上を行う人工知能、ロボットおよび、その機能を実現するソフトウェアに関するものである。
入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、分析結果を使用した検索ができる構造に整理して記録器に記録する。情報の履歴を記録し、記録した履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なプログラムを起動することにより問題解決をする人工知能、ロボットおよび、その機能を実現するソフトウェアに関するものである。また、人工知能の中で構築される知識体系が現実に整合して処理できるかという「記号接地問題」を解決し、処理の過程が人間に理解できるブラックボックス化は排除した「暴走しない」人工知能、ロボットおよび、その機能を実現するソフトウェアに関するものである。
The present invention sequentially constructs a knowledge system autonomously from inputted information, and uses the constructed knowledge system to analyze the situation and identify problems, and to search for information for solving conversations, translations and problems , Artificial intelligence, robots and their functions that start processing programs, generate information necessary for problem solving and incorporate them into knowledge systems, handle problems autonomously and further improve knowledge It relates to software that realizes
Information source, reliability, novelty, field, theme, topic, intention, subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how Why, why, sentence type (normal sentence, question sentence, imperative sentence, conditional sentence, truth, fact, rule, common sense, right and wrong, definition, logic, explanation, hypothesis, prediction, opinion, impression, rumor), formula, Analyze the relationship between physical formulas, chemical formulas, symbols, and related information being recorded, organize them into structures that can be searched using the analysis results, and record them in a recorder. Artificial intelligence, robot, and its functions that record information history, analyze situation, context and problem using the recorded history information, and solve problems by launching programs necessary for problem solving It relates to software that realizes In addition, it solves the "symbol grounding problem" that the knowledge system built in artificial intelligence can be processed in conformity with reality, and eliminates "runaway" artificiality that eliminates the black box that allows humans to understand the processing process It relates to intelligence, robots, and software that realizes their functions.

従来の人工知能技術としてはエキスパートシステム、ニューラルネットワークを使用したものがある。
エキスパートシステムはルール群から構成されるプログラムであり、固定である推論エンジンと可変である知識データベースから構成されている。推論エンジンは規則群を用いた推論を行っており、推論の論理としては命題論理、述語論理、認識論理、様相論理、時相論理、ファジー論理等が規則群として使用されている。知識データベースは一般的に専門家の知識を入力することにより構成するが、定式化してみると規則間で矛盾している等の問題が文献等において指摘されている。また、論理以外の処理を扱うことは困難である。
ニューラルネットワークでは、一般的に入力層、中間層、出力層の人工ニューロンをシリアルに接続し、人工ニューロン間の接続はシナプス学習させることにより実施する。シナプス学習としては誤差逆伝搬法(バックプロパゲーション)が一般的に用いられているが、バックプロパゲーションは下記に示すような欠点が文献等にて指摘されている。
・バックプロパゲーションでは、人工ニューロンで使われる伝達関数が可微分でなければならない。
・バックプロパゲーションによる学習での収斂は非常に遅い。
・バックプロパゲーションによる学習は、必ず収斂するとは限らない。
・広域的な最適解ではなく局所的な誤差最少点に収斂することが多い。

また、ニューラルネットワークは処理の過程がブラックボックスであり、処理の妥当性を確認することが困難である等の欠点が指摘されている。
入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、分析結果を使用した検索ができる構造に整理して記録器に記録する人工知能は無い。また情報の履歴を記録し、記録した履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なプログラムを起動することにより問題解決をする人工知能は無い。さらに、人工知能の中で構築される知識体系が現実に整合して処理できるかという「記号接地問題」を解決し、処理の過程が人間に理解できるブラックボックス化は排除した「暴走しない」人工知能、ロボットおよび、その機能を実現するソフトウェアは従来無い。
Conventional artificial intelligence techniques include those using expert systems and neural networks.
The expert system is a program composed of rules, and is composed of a fixed inference engine and a variable knowledge database. The inference engine performs inference using rule groups, and propositional logic, predicate logic, recognition logic, modal logic, temporal logic, fuzzy logic, etc. are used as rule groups as inference logic. A knowledge database is generally constructed by inputting expert knowledge, but when formulated, problems such as contradictions among rules have been pointed out in the literature. In addition, it is difficult to handle processes other than logic.
In a neural network, artificial neurons in the input layer, intermediate layer, and output layer are generally connected serially, and the connection between the artificial neurons is performed by synaptic learning. As the synapse learning, an error back propagation method (back propagation) is generally used. However, back propagation has the following drawbacks pointed out in the literature.
-In backpropagation, the transfer function used in artificial neurons must be differentiable.
-Convergence in learning by backpropagation is very slow.
・ Back-propagation learning does not always converge.
・ In many cases, it converges to a local minimum error point instead of a global optimal solution.
In addition, it has been pointed out that the neural network has a black box in the process and it is difficult to confirm the validity of the process.
Information source, reliability, novelty, field, theme, topic, intention, subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how Why, why, sentence type (common sentence, question sentence, imperative sentence, conditional sentence, truth, fact, rule, common sense, right and wrong, definition, logic, explanation, hypothesis, prediction, opinion, impression, rumors), formula, There is no artificial intelligence that analyzes the relationship between physical formulas, chemical formulas, symbols, and recorded related information, organizes them into structures that can be searched using the analysis results, and records them in a recorder. Also, there is no artificial intelligence that records information history, analyzes the situation, context, and problem using the recorded history information, and solves the problem by starting a program necessary for problem solving. Furthermore, it solves the “symbol grounding problem” of whether the knowledge system built in artificial intelligence can be processed in conformity with reality, and eliminates the black box that allows humans to understand the processing process. There is no intelligence, robot, or software that realizes its functions.

従来の人工知能技術としてはエキスパートシステム、ニューラルネットワークを使用したものがあるが、いずれも人工知能を実現するためには大きな欠点を有している。
エキスパートシステムは規則群を用いた論理のみが処理の対象である。人間の思考は論理のみで表現することは困難である。また、エキスパートシステムの知識データベースは一般的に専門家の知識を入力することにより構成するが、定式化してみると規則間で矛盾している等の問題が文献等において指摘されている。
ニューラルネットワークでは、一般的に入力層、中間層、出力層の人工ニューロンをシリアルに接続し、人工ニューロン間の接続はシナプス学習させることにより実施する。シナプス学習としては誤差逆伝搬法(バックプロパゲーション)が一般的に用いられているが、バックプロパゲーションは下記に示すような欠点が文献等にて指摘されている。
・バックプロパゲーションでは、人工ニューロンで使われる伝達関数が可微分でなければならない。
・バックプロパゲーションによる学習での収斂は非常に遅い。
・バックプロパゲーションによる学習は、必ず収斂するとは限らない。
・広域的な最適解ではなく局所的な誤差最少点に収斂することが多い。

また、ニューラルネットワークは処理の過程がブラックボックスであり、処理の妥当性を確認することが困難である等の欠点が指摘されている。
人間の思考パターンの遷移のように論理以外の遷移も実施できる人工知能を実現する。また、人工知能の中で構築される知識体系が現実に整合して処理できるかという「記号接地問題」を解決し、処理の過程が人間に理解できるブラックボックス化は排除した「暴走しない」人工知能を実現する。
Conventional artificial intelligence techniques include those using expert systems and neural networks, both of which have major drawbacks for realizing artificial intelligence.
In the expert system, only logic using rules is processed. It is difficult to express human thought only by logic. In addition, the expert system knowledge database is generally configured by inputting expert knowledge. However, when formulated, problems such as contradictions among rules have been pointed out in the literature.
In a neural network, artificial neurons in the input layer, intermediate layer, and output layer are generally connected serially, and the connection between the artificial neurons is performed by synaptic learning. As the synapse learning, an error back propagation method (back propagation) is generally used. However, back propagation has the following drawbacks pointed out in the literature.
-In backpropagation, the transfer function used in artificial neurons must be differentiable.
-Convergence in learning by backpropagation is very slow.
・ Back-propagation learning does not always converge.
・ In many cases, it converges to a local minimum error point instead of a global optimal solution.
In addition, it has been pointed out that the neural network has a black box in the process and it is difficult to confirm the validity of the process.
Realize artificial intelligence that can perform transitions other than logic, such as transitions in human thought patterns. In addition, it solves the "symbol grounding problem" that the knowledge system built in artificial intelligence can be processed in conformity with reality, and eliminates "runaway" artificiality that eliminates the black box that allows humans to understand the processing process Realize intelligence.

人間の思考は言語、数、数式、記号等により表現される。また、言語は単語の組合せにより表現されるので、人間の思考は単語、数、数式、記号の組合せで表現することができる。単語、数、数式、記号をそれぞれに対応するパターンに変換すると、人間の個々の思考は個々のパターンの組合せとして表現することができ、人間の思考の変遷は単語、数、数式、記号等を変換したパターンの組合せからパターンの組合せへの変化としてとらえることができる。
パターンは文および文章のように概念を表現することが可能である。また、パターンは逐次、関連するパターンを起動し、起動したパターンに数々の処理を実行させることが可能である。パターンの起動をトリガーとして処理プログラムを起動し、その結果を別のパターンに取り込むことも可能である。さらに、画像情報、情報の処理、画像パターンから画像パターンへの遷移、画像パターンから言語パターンへの遷移、音声パターンから言語パターンへの遷移および動作を行うために駆動装置を駆動するための信号を生成することも可能である等、扱うことができる範囲が非常に広い概念である。
単語、数、数式、記号に対応するパターンを生成すると、様々な概念はこれらのパターンの組合せで表現することができる。単語、数、数式、記号からパターンへの変換は識別できるものであれば特に形式に制約は無い。なおパターンの組合せも、パターンである。本発明の実施例では各単語に識別番号を割り当てることにより実施した。(例えば「私」という単語の単語識別番号は「0」、助詞の「は」の単語識別番号は「1」、「先生」「2」、・・・「立方体」「101」、「球」「102」、「密度」「103」、「重さ」「104」・・・「人工知能」「169」、「職業」「170」・・・等々)新規の単語が発生した場合には識別番号を逐次追加することにより対応することができる。数々の思考・概念はパターンの組合せ(実施例では各単語、記号に対応付けした識別番号の組合せを主語、述語、修飾関係を識別した構造に整理している)で表現することができる。
本人工知能装置は人間の思考または情報をパターンとして表現し、パターンからパターンへ自律的に遷移させることが可能である。記録ユニットは情報に関する各種の内容を記録している。記録する内容の例としては、情報の内容を表現するパターン、情報間の関係を表現する他の記録ユニットとの接続情報(接続数、接続先、接続先との関係性)、記録ユニットが起動するための条件(照合条件、合致条件)、パターンの分析結果(情報の源泉、信頼性、新規性、分野、テーマ、文の種類)等がある。
文の種類としては、平常文、疑問文、命令文、条件文、感嘆文、真実、事実、規則、常識、定義、論理、説明、仮説、予測、意見、感想、噂、相対的意味を有する単語の絶対化、数式、化学式、記号等を識別し分析結果として記録する。また、情報の内容であるパターンは主語、主語の修飾、述語、述語の修飾、修飾関係、文の種類、数式、物理式、化学式、記号、いつ、どこで、誰が、何を、どのように、なぜ、したのか、および記録している関連情報との関係を分析し、情報を検索し易い構造に整理して記録する。情報が入力されると情報の分野、テーマ、文の種類、構文(主語、主語の修飾、述語、述語の修飾、修飾関係、文の種類、数式、物理式、化学式、記号、いつ、どこで、誰が、何を、どのように、なぜ、したのか)を分析し、記録している情報との照合を行い情報の価値(信頼性、新規性、妥当性、整合性、善悪、既知の情報との関係)を評価する。評価した結果、安全で価値のある情報は分野、テーマ、既知の情報との関係を整理して記録器に記録する。
情報間の関係としては原因と結果、事象と理由、説明と結論、概略と詳細、類似の意味、反対の意見、本文と要約、本文と詳細情報等、様々なものがある。これらの関係性の設定は人間の指示または情報と情報の関係を学習することによる自律設定により実施する。
このように情報は関連情報と関係性を有した知識体系として記録されるので、ある情報が入力され、該当する記録ユニットが起動すると、起動した記録ユニットから関連する記録ユニットの情報(接続先、接続先との関係性)を呼び出し、その中から適切な記録ユニットを逐次、起動し必要な情報を検索していくことが可能となる。記録ユニットの起動をモニターすることにより人工知能装置における処理の中心が現在どこにあるかを確認することができる。人間が思考する場合において、思考または意識の中心が現在何かを確認することに相当する。また、従来型のプログラムで例えればコンパイルした命令コードの実行ポインタの位置をプログラム実行中に逐次、表示することに相当する。
従来の人工知能では情報に含まれる単語を検索キーとして関連情報を検索し、欲しい情報との相関が大きいものを選択するという手法が主として取られている。従って、検索された関連情報が誤ったものであっても、検索キーと相関が大きいものは回答として出力されるという欠点がある。本発明では入力情報を既に評価済みの知識体系と妥当性、整合性を評価した後に知識体系に組み入れていくので、人間が教育を受けながら正しく知能を向上させていくのと同様に誤った情報を排除しつつ知識体系を構築していくことができる。構築している知識体系には単語間の意味関係(同じ意味、反対の意味等)についても識別することができるので、誤った情報、規則、道徳に反する情報は仮に入力されても内部に構築した知識体系と整合性が取れないということを検出し、知識体系に組み込まずに廃棄するので安全である。
また従来の人工知能技術(データ検索、ニューラルネットワーク)では人工知能が出力した結果の妥当性および安全性を詳細に評価することは困難である。なぜならば、データ検索では複雑な検索アルゴリズムを介して相関計算等が実施され結果が導かれるため、人間が理解できる形式および論理で途中状況をトレースするのが難しいからである。またニューラルネットワークも多数の人工ニューロン間の結合を機械学習で強化していくことにより出力結果を生成するが、学習の過程を人間が理解できる形式、論理または言語でトレースするのが困難である。つまり、従来の人工知能技術(データ検索、ニューラルネットワーク)では処理の過程がブラックボックスであるという致命的な欠点を有している。処理過程がブラックボックスであると人工知能の自律化、機能、性能を高めていく段階で人工知能が人間の道徳、規範に反していない(暴走しない)ことを確認できないため非常に危険である。なぜならば人間の道徳、規範に反した知能が偶発的に発生(暴走)しても検出し、除去することができないからである。また人工知能が出力する処理結果が内部に偶発的に発生した人間の道徳、規範に反する誤った知能に基づいて生成されたとしても処理結果の妥当性および安全性を確認、評価できないからである。
一方、本発明では各記録ユニットの情報(照合および合致(起動)の条件、パターン、パターン間の接続関係、接続関係のあるパターンとの関係性およびパターンの分析結果)は全て人間が理解できる言語、記号等をパターン化して表現したものである。また各記録ユニットと接続関係のある記録ユニットおよび接続の関係性についても明確であるので論理展開についても容易にトレースすることができる。本発明による人工知能装置は全ての処理過程がモニター可能であり、かつ人間が理解可能な形式、構造を有した透明性の高いシステムである。つまり本発明による人工知能装置は処理過程の妥当性および安全性を人間が理解し、確認できるシステムとなっている。また、本人工知能装置内に構築する知識体系は真実、事実に基づき人間の道徳、規範に反していないことを逐次、確認しながら構築していく。人間の道徳、規範に反したものは逐次、検出し排除するので誤った知識(人間の道徳、規範に反した知識)は組み込まれない。人工知能の自律化、機能、性能を高めていく段階で人間が設定した規則、常識、道徳に関する知識を逸脱しないことを管理することができる。
内部に構築する知識体系は真実、事実、規則、常識、定義、論理に基づいて構築し、これらと整合しない情報は知識体系に取り込まない。全処理は人間が理解できる言語または記号に対応したもので表示することができるので透明性のあるシステムとなっている。また人工知能の中で構築される知識体系が現実に整合して処理できるかという「記号接地問題」に関しては、本人工知能装置の知識体系は人間が教示する現実に対応する「言語、文、文章、記号、数式、物理式、化学式」を源泉データとし、それを逐次、パターンに変換し、情報間の関係をパターン間の接続関係として構造化したものであるので、本質的に入力した情報に対応する「現実」と等価であり、整合性が取れたものとなっている。
本人工知能装置は人間が築いてきた知性、知識を言語、文、文章等の情報として逐次、入力・評価していき、安全で価値があると判断した情報を既に記録している知識との関係を分析して知識体系に自律的に組み込んでいく。入力情報を既に知識体系に記録している情報と比較・評価し、安全で価値があると判断した情報に分析結果を付加し、さらに発展させるという方式で、知識体系を自己組織的に構築していく。このように構築した知識体系は人間の知性に相当するデータベースとなっており、問題の解決および人間との会話に使用することができる。
本人工知能装置は構築した知識体系を活用し、与えられた問題に関する関連データを検索する。検索された関連データと関係を有する情報に関しても、接続先、関係性を識別することができる。処理の過程(考え方のプロセス)において適切な関係性のものを逐次、選択していくことにより、複数の関連情報から適切なものに絞り込み、思考プロセスを適切な目標に向けて自律的に進めていくことが可能である。本人工知能装置は真実、事実、規則、常識、定義、論理に相当する情報を逐次入力していくことで、情報と情報の接続関係を自律的に生成していく。情報を入力すると情報に対応する数々のパターンが起動し、パターンの集合(概念)とパターンの集合(概念)との間の接続関係が人間の指示または学習により強化していく。文、文章は単独で存在するのではなく、互いに関係を有しており文脈(コンテキスト)の中で位置づけされる。後に出現する文は前に出現した文の影響を受けている。この事を反映するメカニズムとして本人工知能装置は入力した文に応じて対応するパターンが起動し、起動したパターンの接続関係記録部には、当パターンが起動する前に起動したパターンの履歴が転写され、関連するパターンとの接続関係が逐次強化する機能を組み込んでいる。この機能により、文に対応するパターンは全て関連する文との関係、つまり文脈(コンテキスト)を踏まえて起動するように学習が進んでいく動作を実現することができる。つまり、多数の文、文章を入力することにより文脈(コンテキスト)のネットワークが自律的に生成され、ある概念(文、文章)が入力されると、その概念は生成された文脈(コンテキスト)に沿って遷移していくことになる。
このような情報間の接続関係を真実、事実、規則、常識、定義、論理に相当する情報について実施していくと、真実、事実、規則、常識、定義、論理に相当する知識体系を自律的に生成することができる。学習が進むと本人工知能は入力された情報が既に記録している情報と整合しているが、整合していないかの識別が可能となる。つまり、真実、事実、規則、常識、定義、論理に相当する知識、知性を獲得したことになる。情報から情報への遷移は生成した文脈(コンテキスト)に沿って実施することも可能であるし、情報に含まれたキーワードを使用して連想記憶的に情報を検索することにより思考パターンの遷移を実現することも可能である。思考パターンの多様な遷移方法を組み合わせることにより人間に近い会話を実施することもできる。
人間の指示により情報(パターン)間の接続関係を生成する場合はある概念(文、文章)に対応するパターンを起動し、つぎの概念(文、文章)のパターンを起動することにより実施することができる。先に起動したパターンから後で起動したパターンに向け、遷移するように接続関係を生成するように設定しておけば、概念と概念の間の接続関係を逐次、生成していき、次に同様の概念が起動すると、生成した接続関係に沿ってパターンが自動的に遷移するようになる。遷移先が複数の場合は、状況に適した関係性を人間の指示または自律的に設定し、複数の遷移先の候補から状況に適した関係性のものが選択されるようにする。なお、関係性の自律的設定は思考パターンの遷移の目的、状況を検出することにより適切な関係性を設定しておく機能により実施できる。
本発明における基本的な処理(パターンの検索、照合、比較、差異抽出、記録、出力、パターン間の演算処理等)についてはプログラミング言語での実装が容易であり効率的であると考える。個々の思考に対応する個別処理に対応する箇所についてはパターンを分析し分析結果に応じて制御を行うデータ駆動型のシステムが個々にプログラミングすることが不要であるため、柔軟性が高いシステムであると考える。このことから、本人工知能装置は入力した情報(言語)を評価し、評価結果に応じて処理を進めていくシステムとしている。起動する記録ユニットの内容を分析し、対応する処理を実施する。状況に応じて起動する記録ユニットを変更することができるので、数々の処理を非常に柔軟に構成することができる。当然、通常のプログラムのように決まったシーケンスで処理を逐次、実施していくことも可能である。ある問題を解決するための手順を条件付き処理として記録ユニットに記録すると、一連の手順は条件付き処理のグループとして記録され、条件付き処理に対応する記録ユニットのグループが構成されることになる。条件に応じて記録ユニットが起動すると、その記録ユニットに定義された条件の成立性を確認しながら次の処理に対応する記録ユニットを起動していき、一連の手順が実行されていく。
人間がこれまでに蓄積してきた知識を言語等で入力することにより人工知能装置内に知識体系を構築するとともに構築した知識体系を使って問題を解決し、必要に応じて人間が開発してきてきたプログラムを自律的に起動させ、処理結果を活用することができるので、問題解決にあたり人間の柔軟性とコンピュータの処理能力を併せ持つ人工知能を実現することができる。処理の内容をプログラムし、本人工知能装置の中に組み込む、または計算機の中に実装されているプログラムを本人工知能装置とインタフェースを取り本人工知能装置から駆動するという本人工知能装置の柔軟性と機能および計算機の処理能力を兼ね合わせたシステムを構成することも可能である。
本発明の方式では処理の実行中に処理の変更、追加等を容易に実施することが可能である。通常、プログラム言語によるプログラミング処理の場合、処理の変更を実施する場合は該当する箇所のプログラム再設計、再プログラムのインストールおよび再コンパイルが必要であるが、本発明では処理の変更、追加等は該当するパターンとパターンの接続情報をローカルに変更することにより実施できるので、再コンパイルのようにシステム全体を再構築する必要は無い。また、プログラミング言語では、プログラムで使用する変数、定数、関数等の名称は厳密に指定しないと動作しない。これに対して本発明では個々の情報を自然言語に対応するパターンとして表現し、情報の検索に関してはパターンの組合せにより様々な形での検索が可能なシステムである。多少の曖味さがあっても関連する単語の組合せで情報を検索することが可能である。(情報の検索は自然言語に対応したパターンの組合せを使い、同じ意味の文の検出、新規性の検出、記録している情報との整合性確認、相違点抽出等、数々の検索および処理が可能である。)
本発明では処理に使用する変数および定数に対応する情報の検索は自然言語(単語、記号)を組み合わせたパターンを使用することにより実施しており、より人間の思考に近い情報の検索方式となっている。所謂、連想記憶システムにおける情報の記録および検索の機能も有したシステムとなっている。新規の情報が発生した段階で自然言語(単語、記号)の組み合わせたパターンを検索キーとして逐次、登録・記録することができる。また登録・記録した情報を自然言語(単語、記号)の組み合わせた検索パターンで容易に検索することができる。これによりプログラミングにおける使用する変数および定数を事前に厳密に定義する必要があるという制約に縛られない柔軟性が高いシステムを構成することができる。
さらに、処理の内容については条件付き処理を状況(パターンおよびパターンの分析結果により条件の成立性を識別する)に応じて条件に対応する処理を実施し、実施した処理において、さらに条件付き処理が出現すれば、その条件を分析し処理を決定するという方式で逐次処理を進めていくことが可能である。従って、人間が問題解決するのと同じ思考プロセス(条件文の集合)を自然言語で定義することにより、本人工知能装置はその内容を逐次、解釈し必要な処理を進めていくことができる。処理の結果、新しく生成される情報は処理に応じて適切な単語の組合せを検索キーとして記録するので、別の処理において生成した情報を使用する場合でも関連する単語の組合せで検索キーを生成し、容易に検索することができる。つまり生成した情報、知識を知識体系に組み込み、将来において必要になった場合に活用することが容易にできるシステムとなっている。通常のプログラミングでは使用する変数、定数および処理の全てについて事前に定義しておく必要があるが、本発明では処理の結果生成した新規の情報を逐次、知識体系に組み込み、別の処理において知識体系に組み込まれた情報を適宜、欲しい時に検索して処理を進めていくことができる。また情報の検索対象も知識体系全体、直近に処理した処理結果に限定する等、処理に応じて適切に選択することが可能である。
以上では本発明の特徴について説明したが、本人工知能装置に従来型の処理プログラムを組み込み、本人工知能装置の柔軟性(自然言語の解釈、構築した知識体系の活用、自律機能、学習機能等)を有しつつ指定した処理を逐次実施していくというハイブリッドなシステムも当然のことながら実施可能である。
本発明の人工知能装置は従来の人工知能、ロボット等で採用しているエキスパートシステム、有限オートマトン、ニューラルネットワーク、リレーショナルデータベース、仮想テーブルおよび人工知能用のプログラミング言語の機能をカバーするのみならず、それらが有していない機能および柔軟性(完全データ駆動システム、機能拡張の柔軟性(局所的な変更で機能追加可能、連想記憶システム)、曖昧さを許容するシステムへの対応等、無意識(複数の状態)の模擬機能、並行処理への対応、変数および定数の厳密な定義が不要であること、自然言語ベースによる情報の検索が可能であること、条件付き処理の集合を定義することで様々な自律的処理が実施できること、処理のブラックボックス化を排除していること、接続の多様性(フィードフォワード、フィードバック、多対多結合等)、可微分性の制約が無いこと、バックプロパゲーションにおける収斂の問題が無い等)を有したシステムとなっている。
人間の思考は大きく2つに分類することができる。一つは思考の内容であり、もう一つは思考の仕方(手順、プロセス、考え方)である。本発明の実施例では思考の内容に関しては自然言語を変換したパターンとして定義し、思考の仕方に相当する箇所をプログラムで処理方法を設定することにより機能を実現している。思考の内容は多種多様であるが、思考の仕方は共通的、一般的、汎用的である。例えば、入力情報の処理としては文の種類(平常文、疑問文、命令文、条件文)に応じて対応する処理を設定することができる。平常文であれば入力情報の信頼性、妥当性を評価し、安全で価値が有ると評価すれば知識体系に組み入れ知識体系の向上を図る。疑問文であれば記録している知識体系から関連情報を検索し、適切な情報があれば回答として出力する。命令文であれば、過去に同様の命令を実施した経験の有無、実施手順、実施した場合の影響(人間の道徳および規範を逸脱しないか等を含む)を評価し対応を決定する。条件文であれば、条件に相当する内容の成立性を知識体系または外部情報から分析し、分析結果に応じた処理を行う。以上に述べた思考の仕方に相当する部分をプログラムで処理方法を実装することにより、多種多様な思考の内容は人間がこれまでに蓄積してきた様々な知識を知識体系の源泉として自然言語のままで処理することが可能なシステムを構成することができる。従って、非常に柔軟性、拡張性のある人工知能が非常に小さな労力、開発コストで実現することが可能である。(従来の人工知能では思考の内容に相当する部分も逐一プログラムで実装する必要があり各機能の実現のためには大きな開発コストを要する。)本発明の記録ユニットは情報(思考、概念)から情報(思考、概念)への遷移を情報間の関係性を使用することで実施できる。人間は思考する際に、思考内容をどのように変化させるかを関係性(原因と結果、事象と理由、説明と結論、概略と詳細、類似内容、反対の意見等)を使用して実施しており、本人工知能装置も情報(パターン)から情報(パターン)への遷移をパターン間の関係性を使用して遷移させることが可能である。ある情報が入力された場合、関連情報は多岐にわたるが、関連情報の内、各思考段階において適切な関係性(原因と結果、事象と理由、説明と結論、概略と詳細、類似内容、反対の意見等)を選択することにより思考のステップを思考の仕方に沿って適切な次ステップに移行させることができる。
記録ユニットの起動方法としては大きく分けて3通りある。一つ目はパターン制御器から検索用のパターンを各記録ユニットに照射し、相関のある記録ユニットを起動する方法である。(ソフトウェアでの実現方法としては検索用のパターンと各記録ユニットに記録しているパターンとの照合検索を実施する方法がある。)二つ目はパターン照射器から各記録ユニットの接続情報記録部に対し設定した期間分の起動履歴を照射し、起動条件を満足した記録ユニットを起動する方法である。(ソフトウェアでの実現方法としては起動履歴から一定期間分の履歴パターンを切り出し、各記録ユニットの接続情報記録部に記録しているパターンと照合し、起動条件を満足する記録ユニットを抽出する方法がある。)三つ目は記録ユニットに記録している関連する記録ユニットとの接続情報(接続数、接続先、接続先との関係性)を使い、適切な関係性を有する記録ユニットを起動する方法である。(ソフトウェでの実現方法としては選択された記録ユニットに記録している関連する記録ユニットとの接続情報(接続数、接続先、接続先との関係性)を読出し、次ステップの記録ユニットとして適切な(相応しい)関係性のものを抽出し、適切な関係性を有した記録ユニットの接続先を選択し、起動する方法がある。)本人工知能装置ではこれら3通りの起動方法を各処理において適宜活用することにより、思考パターンを遷移させる上で最も適切な記録ユニットを起動するようにしている。記録ユニットの起動をモニターすることにより人工知能装置における処理の中心が現在どこにあるかを確認することができる。これは人間が思考する場合において、思考および概念の中心が現在何かを確認することに相当する。
本人工知能装置は自律学習の機能も有している。外部から入力した情報により該当する記録ユニットは逐次起動されていくが、この起動の履歴はパターン照射器に記録される。各記録ユニットが起動する毎に、起動した記録ユニットの接続情報記録部に起動履歴の一部(起動した時点から過去にさかのぼった一定期間の起動履歴)を記録することにより、起動した記録ユニットと起動に関係が強いと考えられる記録ユニットとの接続関係を逐次、強化することができる。入力情報が学習データである場合、学習データを逐次入力していくことにより、学習データ間の関係(論理関係、因果関係、時系列関係等)が強化される。学習完了後は一部の情報を入力すると、その情報に関連する学習データの記録ユニットが記録されている接続関係に従って逐次起動し、学習した内容が自動的に想起される。なお、各ステップにおいて複数の記録ユニットが同時に起動することが可能である。
一連の文を逐次、入力していくと文をパターンに変換してパターン記録器の各記録ユニットに記録していく。文と文の関係性(原因と結果、事象と理由、説明と結論、概略と詳細、類似内容、反対の意見等)について識別できるものは各関係性に対応する記録ユニットを人間の指示または関係性の学習による自動設定により起動する。このように文と文との関係性が識別される度に関係性を表現する記録ユニットも起動していくと、関係性を示す記録ユニットの接続情報記録部に関係性を有した文と文の起動履歴が転写され接続関係が強化されていく。多数の文と文を入力することにより様々な関係性のタイプを学習していくことになる。初期段階において人間が文と文の関係性について教示(関係性が識別された段階で関係性を表現する記録ユニットを人間の指示により起動することで実施)していくと本人工知能装置は関係性の識別方法について学習することになる。
関係性は多対多で結合されたパターン間の接続関係から各思考段階において適切な接続関係を選択する際に活用することができる。思考を結果の方向に遷移させ最終的な結果を知りたいのか、それとも思考を原因の方向に遷移させ事象が生じた根本原因を知りたいのかで思考の遷移の仕方は異なる。関係性に何を選択するかで遷移の方向を制御することができる。
前述の説明において文と文の関係性を人間の指示または学習により識別することが可能であることを示した。識別した関係性は各記録ユニットの他の記録ユニットとの接続関係(接続数、接続先、関係性)を記録している箇所に逐次、記録していく。このように一連の文を入力していくと文は単独で記録するのではなく文と文の関係性を含めて記録することになる。
本人工知能装置では入力情報の処理を個々にプログラムするのではなく、機械に情報と情報の関係を分析する方法、知識体系の構築の方法および問題解決の方法、入力した情報を一般化する方法等、入力情報の処理の方法を実装することにより実施することが可能である。(当然のことながら入力した情報の処理に関して、個々にプログラムして本人工知能装置に実施させることも可能である。)言語、数、数式、記号で表現された情報を解釈し、解釈した結果に応じて適切な処理を自律的に行う。従来では処理の内容はプログラムで表現されていたが、本発明では自然言語で表現された関連情報に対応するパターンを検索し、条件との照合、条件に応じた処理および分岐を実施することにより次ステップの処理を自律的に行うので、個々の処理の内容をプログラムする必要はなく、自然言語で処理の内容を指示することで済む。つまり各処理を逐次プログラムしていく必要は無く、既に自然言語で表されている知識を本人工知能装置の知識体系の源泉として活用することができる。(本人工知能装置は自然言語で入力した知識を逐次、対応するパターンへ変換して情報と情報の関係性を分析し、さらに整理して知識体系に組み込んでいく。)当然のことながら、各処理を逐次プログラムしたものを記録ユニットとして定義し、条件照合により条件が合致した場合に定義した記録ユニットを起動することにより動作させても良い。
入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、主語、主語の修飾、述語、述語の修飾、修飾関係、文の種類、数式、化学式、記号、いつ、どこで、誰が、何を、どのように、なぜ、したのか、および記録している関連情報との関係を分析し情報を検索し易い構造に整理して記録するとともに、真理、真実、事実、専門知識、規則および常識を情報と情報の間に関係性を有した知識体系として構築する。
初期では人間が処理の方法を教示し、機械は人間から教示された方法を記録、学習していく。ある程度、学習が進むと機械は自律的に処理を実行するようになる。機械の処理結果を人間が確認し、処理が誤っていれば機械に通知し、処理の修正を適宜行う。学習はパターンとパターンとの関係を指定することにより実施する。自然言語で表現した文と文の関係を指定することにより機械は自動的にパターンとパターンとの関係として接続関係を生成していく。人間の思考プロセスは一般的に思考の内容と思考の仕方に区別することができる。思考の内容は多種多様であるが、思考の仕方は一般的、普遍的、共通的な特徴を有している。本発明では思考の内容に対応するパターンを思考の仕方に対応するパターンまたはプログラムで制御することにより思考のプロセスを適切かつ有意な方向に遷移させていく。情報と情報の関係も一対一の関係ではなく、多対多の関係である。ただし、その関係は一般化して対応づけることができる。例えば関係性は原因と結果、事象と理由、説明と結論、概略と詳細等のように識別することができる。思考パターンの遷移においては、思考段階に応じて各種の思考の仕方および情報と情報の関係の内、適切なものを選択していくことにより、効率的に思考パターンを目標に到達させることができる。このように従来では多種多様の処理を個別にプログラムする必要があったが、多種多様な処理に対応する部分は自然言語(文、文章)に対応するパターンおよびパターン間の遷移で設定し、思考の仕方および情報間の関係を選択する機能に関する部分をプログラムで実装することにより、大幅に開発に要する労力を削減することができる。なお、思考の仕方および情報間の関係の選択する機能に関してもプログラムでの実装ではなくパターン間の処理で実現することも可能である。
Human thought is expressed by language, numbers, mathematical formulas, symbols, and the like. In addition, since a language is expressed by a combination of words, human thought can be expressed by a combination of words, numbers, mathematical expressions, and symbols. When words, numbers, formulas, and symbols are converted into corresponding patterns, each human thought can be expressed as a combination of individual patterns. It can be regarded as a change from a combination of converted patterns to a combination of patterns.
Patterns can express concepts like sentences and sentences. In addition, it is possible to sequentially activate related patterns and cause the activated patterns to execute a number of processes. It is also possible to start a processing program using a pattern start as a trigger and load the result into another pattern. In addition, image information, information processing, transition from image pattern to image pattern, transition from image pattern to language pattern, transition from voice pattern to language pattern, and signals for driving the driving device to perform operations It is a concept with a very wide range that can be handled, such as being able to be generated.
When patterns corresponding to words, numbers, mathematical expressions, and symbols are generated, various concepts can be expressed by combinations of these patterns. There are no restrictions on the format of words, numbers, mathematical expressions, and conversion from symbols to patterns as long as they can be identified. A combination of patterns is also a pattern. In the embodiment of the present invention, an identification number is assigned to each word. (For example, the word identification number of the word “I” is “0”, the word identification number of the particle “ha” is “1”, “teacher” “2”,... “Cube” “101”, “sphere” “102”, “density” “103”, “weight” “104”... “Artificial intelligence” “169”, “occupation” “170”. This can be dealt with by sequentially adding numbers. Numerous thoughts / concepts can be expressed by combinations of patterns (in the embodiment, combinations of identification numbers associated with words and symbols are arranged in a structure in which the subject, predicate, and modification relationship are identified).
This artificial intelligence device can express human thought or information as a pattern and can autonomously transition from pattern to pattern. The recording unit records various contents related to information. Examples of contents to be recorded include patterns that express information contents, connection information with other recording units that express relationships between information (number of connections, connection destinations, relationships with connection destinations), and recording unit activation Conditions (matching conditions, matching conditions), pattern analysis results (information source, reliability, novelty, field, theme, sentence type), and the like.
The types of sentences include ordinary sentences, question sentences, imperative sentences, conditional sentences, exclamation sentences, truth, facts, rules, common sense, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors, and relative meanings. Identify word absoluteization, mathematical formulas, chemical formulas, symbols, etc. and record them as analysis results. In addition, the pattern that is the content of the information is the subject, subject modification, predicate, predicate modification, modification relationship, sentence type, formula, physical formula, chemical formula, symbol, when, where, who, what, how, Analyze why and why, and the relationship with the recorded related information, and organize and record the information in a structure that is easy to search. When information is entered, the field of information, theme, sentence type, syntax (subject, subject modification, predicate, predicate modification, modification relationship, sentence type, formula, physical formula, chemical formula, symbol, when, where, Analyze who, what, how and why) and compare it with the recorded information to determine the value of the information (reliability, novelty, validity, integrity, good / bad, known information) The relationship). As a result of the evaluation, safe and valuable information is recorded in a recorder by organizing the relationship with the field, theme, and known information.
There are various relationships between information such as causes and results, events and reasons, explanations and conclusions, outlines and details, similar meanings, opposite opinions, texts and summaries, texts and details. Setting of these relationships is performed by an autonomous setting by learning a human instruction or a relationship between information and information.
Since information is recorded as a knowledge system having a relationship with related information in this way, when certain information is input and the corresponding recording unit is activated, information about the associated recording unit (connection destination, (Relationship with connection destination) can be called, and appropriate recording units can be sequentially activated from among them to search for necessary information. By monitoring the activation of the recording unit, it is possible to confirm where the processing center in the artificial intelligence apparatus is currently located. In human thinking, this is equivalent to confirming what is currently the center of thought or consciousness. For example, in the case of a conventional program, this corresponds to sequentially displaying the position of the execution pointer of the compiled instruction code during program execution.
Conventional artificial intelligence mainly takes a technique of searching related information using a word included in information as a search key and selecting one having a large correlation with desired information. Therefore, even if the retrieved related information is incorrect, there is a disadvantage that information having a large correlation with the retrieval key is output as an answer. In the present invention, the input information is incorporated into the knowledge system after evaluating the validity and consistency with the already evaluated knowledge system, so that erroneous information is just as well as improving human intelligence while receiving education. It is possible to build a knowledge system while eliminating The built knowledge system can also identify semantic relationships between words (same meanings, opposite meanings, etc.), so if wrong information, rules, or information that violates morals is entered, it is built internally It is safe because it detects that it is not consistent with the knowledge system and discards it without incorporating it into the knowledge system.
In addition, it is difficult to evaluate in detail the validity and safety of the results output by artificial intelligence using conventional artificial intelligence technology (data retrieval, neural network). This is because, in data retrieval, correlation calculation and the like are performed through a complicated retrieval algorithm and the result is derived, so it is difficult to trace the intermediate situation in a format and logic that can be understood by humans. A neural network also generates output results by strengthening the connection between a large number of artificial neurons by machine learning, but it is difficult to trace the learning process in a form, logic or language that can be understood by humans. That is, the conventional artificial intelligence technology (data search, neural network) has a fatal defect that the process is a black box. If the process is a black box, it is very dangerous because it cannot be confirmed that the artificial intelligence does not violate human morals and norms (does not run away) at the stage of increasing the autonomy, function, and performance of artificial intelligence. This is because even if human morals and norms of intelligence are accidentally generated (runaway), they cannot be detected and removed. In addition, even if the processing result output by artificial intelligence is generated based on human morality or accidental intelligence that occurs accidentally inside, the validity and safety of the processing result cannot be confirmed and evaluated. .
On the other hand, according to the present invention, all the information of each recording unit (matching and matching (activation) conditions, patterns, connection relations between patterns, relations with patterns having connection relations, and analysis results of patterns) can be understood by humans. , Symbols and the like are expressed as patterns. In addition, since the recording unit having connection relation with each recording unit and the relation of connection are clear, the logical development can be easily traced. The artificial intelligence apparatus according to the present invention is a highly transparent system having a format and a structure in which all processing processes can be monitored and can be understood by humans. That is, the artificial intelligence apparatus according to the present invention is a system that allows a human to understand and confirm the validity and safety of the processing process. In addition, the knowledge system built in this artificial intelligence device will be built while successively confirming that it is not against human morals and norms based on truth and facts. Since things that violate human morals and norms are successively detected and eliminated, false knowledge (knowing that violates human morals and norms) is not incorporated. It is possible to manage not to deviate from knowledge about rules, common sense, and morals set by humans at the stage of increasing the autonomy, function, and performance of artificial intelligence.
The knowledge system built inside is based on truth, facts, rules, common sense, definition, and logic, and information that does not match these is not taken into the knowledge system. The entire process can be displayed in a language or symbol that can be understood by humans, thus making the system transparent. In addition, regarding the “symbol grounding problem” that the knowledge system built in artificial intelligence can be processed consistently with the reality, the knowledge system of this artificial intelligence device corresponds to the “language, sentence, Essentially input information because the text, symbols, mathematical formulas, physical formulas, chemical formulas are used as source data, and the data is sequentially converted into patterns, and the relationships between the information are structured as connection relationships between the patterns. It is equivalent to the “reality” corresponding to, and is consistent.
This artificial intelligence device sequentially inputs and evaluates intelligence and knowledge that humans have built as information such as language, sentences, sentences, etc., and knowledge that has already been recorded information that is judged safe and valuable Analyzing relationships and incorporating them into the knowledge system autonomously. Comparing and evaluating input information with information already recorded in the knowledge system, adding analysis results to information judged safe and valuable, and further developing the knowledge system in a self-organizing manner To go. The knowledge system constructed in this way is a database corresponding to human intelligence, and can be used for problem solving and conversation with humans.
This artificial intelligence device uses the constructed knowledge system to search related data related to a given problem. The connection destination and the relationship can also be identified for the information related to the retrieved related data. By selecting the relevant ones in the processing process (thinking process) one after another, the relevant information is narrowed down to the appropriate one, and the thinking process proceeds autonomously toward the appropriate goal. It is possible to go. This artificial intelligence device autonomously generates the connection relationship between information and information by sequentially inputting information corresponding to truth, fact, rule, common sense, definition and logic. When information is input, a number of patterns corresponding to the information are activated, and the connection relationship between the pattern set (concept) and the pattern set (concept) is strengthened by human instruction or learning. Sentences and sentences do not exist alone, but are related to each other and positioned in the context. The sentence that appears later is affected by the sentence that appears earlier. As a mechanism to reflect this, this artificial intelligence device activates the corresponding pattern according to the input sentence, and the history of the pattern activated before this pattern is activated is transferred to the connection relation recording part of the activated pattern. It incorporates a function that strengthens the connection relationship with related patterns. With this function, it is possible to realize an operation in which learning proceeds so that all patterns corresponding to a sentence are activated based on a relationship with a related sentence, that is, a context. In other words, by inputting a large number of sentences and sentences, a context network is generated autonomously. When a certain concept (sentence or sentence) is input, the concept follows the generated context (context). Will make a transition.
When such information is connected to information corresponding to truth, facts, rules, common sense, definition, and logic, the knowledge system corresponding to truth, facts, rules, common sense, definition, and logic is autonomous. Can be generated. As learning proceeds, the artificial intelligence can identify whether the input information is consistent with the information already recorded, but not. In other words, it has acquired knowledge and intelligence equivalent to truth, facts, rules, common sense, definition and logic. The transition from information to information can be carried out along the generated context (context), and the transition of thought patterns can be performed by searching for information in an associative memory using keywords included in the information. It can also be realized. By combining various transition methods of thought patterns, it is possible to carry out a conversation close to a human being.
When a connection relationship between information (patterns) is generated by a human instruction, the pattern corresponding to a certain concept (sentence, sentence) is activated and the pattern of the next concept (sentence, sentence) is activated. Can do. If it is set so that the connection relationship is generated so as to transition from the pattern activated first to the pattern activated later, the connection relationship between concepts will be generated sequentially, and then the same When the concept is activated, the pattern automatically transitions along the generated connection relationship. When there are a plurality of transition destinations, a relationship suitable for the situation is set by a human instruction or autonomously, and a relationship suitable for the situation is selected from a plurality of transition destination candidates. Note that the autonomous setting of the relationship can be performed by a function of setting an appropriate relationship by detecting the purpose and situation of the transition of the thought pattern.
Basic processing (pattern search, collation, comparison, difference extraction, recording, output, arithmetic processing between patterns, etc.) in the present invention is easy to implement in a programming language and is considered efficient. It is a highly flexible system because it is not necessary to individually program a data-driven system that analyzes patterns and performs control according to the analysis results for locations corresponding to individual processes corresponding to individual thoughts. I think. For this reason, the artificial intelligence apparatus evaluates the input information (language) and advances the processing according to the evaluation result. Analyzes the contents of the recording unit to be activated and performs the corresponding processing. Since the recording unit to be activated can be changed according to the situation, many processes can be configured very flexibly. Of course, it is also possible to carry out the processing sequentially in a fixed sequence like a normal program. When a procedure for solving a certain problem is recorded in the recording unit as a conditional process, a series of procedures is recorded as a group of conditional processes, and a group of recording units corresponding to the conditional process is configured. When the recording unit is activated according to the conditions, the recording unit corresponding to the next process is activated while confirming the satisfaction of the condition defined for the recording unit, and a series of procedures is executed.
The knowledge system that humans have accumulated so far is input in language, etc., and the knowledge system is built in the artificial intelligence device, and the problem is solved using the constructed knowledge system, and humans have developed as needed Since the program can be activated autonomously and the processing results can be used, artificial intelligence that combines human flexibility and computer processing power can be realized in solving problems. The flexibility of the artificial intelligence device that programs the contents of processing and incorporates it in the artificial intelligence device, or drives the program implemented in the computer by interfacing with the artificial intelligence device and driving from the artificial intelligence device It is also possible to configure a system that combines the functions and the processing capacity of the computer.
In the method of the present invention, it is possible to easily change or add a process during the execution of the process. Normally, in the case of programming processing in a programming language, it is necessary to redesign the program at the corresponding location, install the reprogram, and recompile when executing the processing change. However, in the present invention, the processing change, addition, etc. are applicable. This can be done by changing the pattern to be connected and the connection information of the pattern locally, so there is no need to rebuild the entire system like recompilation. In programming languages, the names of variables, constants, functions, etc. used in the program will not work unless they are strictly specified. On the other hand, in the present invention, each information is expressed as a pattern corresponding to a natural language, and the information can be searched in various forms by combining patterns. Even if there is some ambiguity, it is possible to search for information by a combination of related words. (Search for information uses a combination of patterns corresponding to natural language, and many searches and processes such as sentence detection with the same meaning, detection of novelty, consistency check with recorded information, extraction of differences, etc. Is possible.)
In the present invention, searching for information corresponding to variables and constants used for processing is performed by using a pattern that combines natural languages (words, symbols), and this is a method for searching for information closer to human thinking. ing. The so-called associative memory system has a function of recording and retrieving information. When new information is generated, a combination pattern of natural languages (words and symbols) can be sequentially registered and recorded as a search key. In addition, the registered / recorded information can be easily searched with a search pattern that combines natural languages (words, symbols). As a result, it is possible to construct a highly flexible system that is not constrained by the restriction that variables and constants to be used in programming must be strictly defined in advance.
Furthermore, with regard to the content of the processing, the processing corresponding to the condition is performed according to the situation (identifying the validity of the condition based on the pattern and the analysis result of the pattern). If they appear, it is possible to proceed sequentially by analyzing the conditions and determining the processing. Therefore, by defining the same thinking process (a set of conditional sentences) that a human being solves a problem in natural language, the artificial intelligence apparatus can sequentially interpret the contents and proceed with necessary processing. As a result of processing, newly generated information records a combination of appropriate words as a search key according to the processing, so even when using information generated in another processing, a search key is generated with a combination of related words. Can be easily searched. In other words, the generated information and knowledge are incorporated into the knowledge system and can be easily utilized when needed in the future. In normal programming, it is necessary to predefine all variables, constants, and processes to be used. However, in the present invention, new information generated as a result of the process is sequentially incorporated into the knowledge system, and the knowledge system in another process. It is possible to search for the information embedded in the information as needed and proceed with the processing. Also, the information search target can be appropriately selected according to the processing, such as limiting the entire knowledge system and the processing result processed most recently.
Although the features of the present invention have been described above, a conventional processing program is incorporated into the artificial intelligence device, and the flexibility of the artificial intelligence device (interpretation of natural language, utilization of the constructed knowledge system, autonomous function, learning function, etc.) Naturally, it is possible to implement a hybrid system in which specified processing is sequentially performed while having the above.
The artificial intelligence apparatus of the present invention not only covers the functions of the conventional artificial intelligence, expert system, finite automaton, neural network, relational database, virtual table and artificial intelligence programming language used in robots, etc. Functions and flexibility that are not possessed (complete data-driven system, flexibility of function expansion (addition of functions by local change, associative memory system), support for systems that allow ambiguity, etc. (State) simulation function, support for parallel processing, strict definition of variables and constants is not required, information search based on natural language is possible, and various conditional processing sets are defined. The ability to perform autonomous processing, the elimination of black box processing, and the diversity of connections ( Over-forward, feedback, many-to-many bond, etc.), it differentiability constraint is not, has a system having a back-propagation is not convergence problems in Deployment etc.).
Human thought can be broadly classified into two. One is the content of thought, and the other is the way of thinking (procedure, process, way of thinking). In the embodiment of the present invention, the content of thought is defined as a pattern obtained by converting natural language, and a function is realized by setting a processing method in a program corresponding to the way of thinking. Thought contents vary widely, the way of thinking is common, general, and general. For example, a process corresponding to the type of sentence (normal sentence, question sentence, command sentence, conditional sentence) can be set as the process of input information. If it is a normal sentence, the reliability and validity of the input information will be evaluated, and if it is safe and valuable, it will be incorporated into the knowledge system to improve the knowledge system. If it is an interrogative sentence, related information is searched from the recorded knowledge system, and if there is appropriate information, it is output as an answer. If it is an imperative sentence, it evaluates the presence or absence of the experience which implemented the same instruction in the past, an execution procedure, and the influence (when it does not deviate from human morality and a norm), and determines a response | compatibility. If it is a conditional statement, the establishment of the content corresponding to the condition is analyzed from the knowledge system or external information, and processing according to the analysis result is performed. By implementing the processing method corresponding to the way of thinking described above with a program, the content of a wide variety of thoughts remains the natural language as the source of the knowledge system using the various knowledge that humans have accumulated so far. It is possible to configure a system that can be processed with Therefore, it is possible to realize a very flexible and expandable artificial intelligence with very little effort and development cost. (In the conventional artificial intelligence, the part corresponding to the content of thinking must be implemented in a program one by one, and a large development cost is required to realize each function.) The recording unit of the present invention is based on information (thinking, concept). Transition to information (thinking, concept) can be implemented by using the relationship between information. When thinking, humans use relationships (causes and consequences, events and reasons, explanations and conclusions, outlines and details, similar content, opposition opinions, etc.) to change how they think. Therefore, this artificial intelligence apparatus can also transition from information (pattern) to information (pattern) using the relationship between patterns. When certain information is entered, the related information is diverse, but the relevant information has an appropriate relationship (cause and result, event and reason, explanation and conclusion, summary and details, similar content, opposite) It is possible to shift the thinking step to an appropriate next step according to the way of thinking by selecting (such as opinion).
There are roughly three ways to start the recording unit. The first method is to irradiate each recording unit with a search pattern from the pattern controller and start a correlated recording unit. (A software implementation method includes a method for performing a collation search between a search pattern and a pattern recorded in each recording unit.) The second is a connection information recording unit of each recording unit from the pattern irradiator. Is a method for activating a recording unit that satisfies the activation conditions by irradiating the activation history for a set period. (A software implementation method is to extract a history pattern for a certain period from the activation history, compare it with the pattern recorded in the connection information recording section of each recording unit, and extract a recording unit that satisfies the activation condition. The third is using the connection information (number of connections, connection destination, relationship with connection destination) with the related recording unit recorded in the recording unit, and starting the recording unit with an appropriate relationship. Is the method. (The software implementation method reads the connection information (number of connections, connection destination, relationship with connection destination) with the related recording unit recorded in the selected recording unit, and as the recording unit for the next step. There is a method of extracting the appropriate (appropriate) relationship, selecting the connection destination of the recording unit having the appropriate relationship, and starting it.) In this artificial intelligence device, these three activation methods are each processed. Is used as appropriate in order to activate the most appropriate recording unit for changing the thinking pattern. By monitoring the activation of the recording unit, it is possible to confirm where the processing center in the artificial intelligence apparatus is currently located. This is equivalent to confirming what is currently the center of thought and concept in the case of human thinking.
This artificial intelligence device also has an autonomous learning function. The corresponding recording unit is sequentially activated by information input from the outside, and this activation history is recorded in the pattern irradiator. When each recording unit is activated, a part of the activation history is recorded in the connection information recording section of the activated recording unit (a certain period of activation history going back to the past from the time of activation). The connection relationship with the recording unit, which is considered to be strongly related to activation, can be strengthened sequentially. When the input information is learning data, the learning data are sequentially input, thereby strengthening the relationship between the learning data (logical relationship, causal relationship, time series relationship, etc.). When a part of information is input after the learning is completed, the learning data recording unit related to the information is sequentially activated according to the recorded connection relation, and the learned content is automatically recalled. In each step, a plurality of recording units can be activated simultaneously.
When a series of sentences are sequentially input, the sentence is converted into a pattern and recorded in each recording unit of the pattern recorder. Those that can identify sentence-to-sentence relationships (causes and consequences, events and reasons, explanations and conclusions, outlines and details, similar contents, opposite opinions, etc.) indicate the recording unit corresponding to each relationship to a human instruction or relationship It starts by automatic setting by sex learning. When the recording unit that expresses the relationship is activated every time the relationship between the sentences is identified in this way, the sentence and the sentence having the relationship with the connection information recording unit of the recording unit that indicates the relationship are activated. The activation history is transcribed and the connection is strengthened. By inputting a large number of sentences and sentences, various types of relationships are learned. When the human teaches the relationship between sentences at the initial stage (executed by starting a recording unit that expresses the relationship at the stage when the relation is identified, by the human instruction), the artificial intelligence device is related. You will learn how to identify sex.
The relationship can be utilized when selecting an appropriate connection relationship at each thought stage from the connection relationship between the many-to-many connected patterns. The way of thinking transitions differs depending on whether you want to change the thought in the direction of the result and know the final result, or whether you want to change the thought in the direction of the cause and know the root cause that caused the event. The direction of transition can be controlled by what is selected for the relationship.
In the above description, it was shown that the relationship between sentences can be identified by human instruction or learning. The identified relationship is sequentially recorded at a location where the connection relationship (number of connections, connection destination, relationship) with each recording unit is recorded. When a series of sentences are input in this way, the sentence is not recorded alone but is recorded including the relationship between the sentences.
Rather than individually programming the processing of input information in this artificial intelligence device, a method for analyzing the relationship between information and information in a machine, a method for constructing a knowledge system, a method for solving problems, and a method for generalizing input information It is possible to implement by implementing a method of processing input information. (As a matter of course, it is possible to individually program the processing of the input information and have the artificial intelligence device implement it.) Interpretation and interpretation of information expressed in language, numbers, mathematical formulas, and symbols Appropriate processing is performed autonomously according to the situation. Conventionally, the contents of processing are expressed by a program, but in the present invention, a pattern corresponding to related information expressed in a natural language is searched, and matching with a condition, processing and branching according to the condition are performed. Since the process of the next step is performed autonomously, it is not necessary to program the contents of each process, and it is sufficient to indicate the contents of the process in a natural language. In other words, it is not necessary to program each process sequentially, and knowledge already expressed in natural language can be used as a source of knowledge system of the artificial intelligence device. (This artificial intelligence device sequentially converts knowledge input in natural language into corresponding patterns to analyze the relationship between information and information, and organizes it into the knowledge system.) It is also possible to define a recording unit that is a program of processing sequentially as a recording unit, and activate the recording unit that is defined when the condition is met by condition matching.
Information source, reliability, novelty, field, theme, subject, subject modification, predicate, predicate modification, modification relationship, sentence type, formula, chemical formula, symbol, when, where, who, what Analyze how, why, and why, and the relationship with the recorded related information, organize the information into a structure that is easy to search, and record truth, truth, facts, expertise, rules, and common sense Is constructed as a knowledge system with a relationship between information.
Initially, a human teaches the processing method, and the machine records and learns the method taught by the human. As learning progresses to some extent, the machine will autonomously execute processing. A human confirms the processing result of the machine, and if the process is incorrect, notifies the machine and corrects the process appropriately. Learning is performed by specifying the relationship between patterns. By specifying the relationship between sentences expressed in natural language, the machine automatically generates a connection relationship as a relationship between patterns. The human thought process can be generally divided into the content of thought and the way of thinking. Thought content varies widely, the way of thinking has general, universal, and common features. In the present invention, a thought process is shifted in an appropriate and significant direction by controlling a pattern corresponding to the content of thought with a pattern or program corresponding to the way of thinking. The relationship between information and information is not a one-to-one relationship but a many-to-many relationship. However, the relationship can be generalized and associated. For example, relationships can be identified as cause and effect, event and reason, explanation and conclusion, summary and details, and so on. In the transition of thought patterns, the thought pattern can be efficiently reached to the target by selecting an appropriate one among various ways of thinking and the relationship between information and information according to the thought stage. . As described above, conventionally, it has been necessary to individually program various kinds of processing. However, a part corresponding to various kinds of processing is set by a pattern corresponding to a natural language (sentence, sentence) and a transition between patterns, and thinking. By implementing the part related to the function of selecting the relationship between the information and the information in the program, the labor required for development can be greatly reduced. Note that the function of selecting the way of thinking and the relationship between information can also be realized by processing between patterns instead of implementation by a program.

人間の思考を表現する言語をパターンに変換し、パターンおよびパターン間の関係を分析すると、文の種類(平常文、疑問文、命令文等)、特徴(善悪、真理、真実、事実、定義、規則、常識、説明、仮説、予測、意見、感想)および関係(原因と結果、事象と理由、説明と結論、概略と詳細等)を識別することができる。文の種類および特徴は人間が識別できるように、機械にも識別を学習させることが可能である。文の種類は文に含まれる単語を分析することにより、平常文、疑問文、命令文、条件文等の識別が可能である。文の特徴のうち、善悪、真理、真実、事実、定義、規則、常識については個々に関して人間が教示して機械に学習させる。これは各情報を機械に入力する際に、個々の情報の種類(真理、真実、事実、定義、規則、常識)についても識別し、付帯情報を付加して記録することにより実施できる。その他の情報の種類(説明、仮説、予測、意見、感想)の識別は文の種類が(真理、真実、事実、定義、規則、常識)として記録されていないことと、文に含まれる単語(だろう、考える、思う・・等)を分析することにより実施することができる。
言語をパターンに変換し、文を構成する単語に対応するパターンの集合として表現し、単語間の意味的な関係をパターンとパターンとの接続関係で表現すると、文のもつ意味をパターンとパターンの接続関係で表現することができる。入力した情報は語列から単語の識別、単語の特徴(品詞、意味)、文要素の分析、文の構造分析(主語、述語、主語の修飾、述語の修飾、修飾の関係)、文要素と文要素の関係分析(同じ意味、定義、反対の意味等)が実施され、文要素と文要素の関係をパターンとパターンの接続関係として対応付けることにより、情報と情報の意味の関係をパターンとパターンの関係と構造に変換していく。パターンとパターンの関係は文要素と文要素だけでなく、文と文、文章と文章の関係についても設定することができる。これは、パターンをグループ化したものを新たなパターンとして定義することにより実施できる。パターンとパターンの関係(原因と結果、事象と理由、説明と結論、概略と詳細等)を示す特徴的な単語を検出すると、該当するパターン間に対応する関係を自律的に記録する。この記録されたパターン間の関係はパターン間の遷移を制御する際に活用することができる。
パターンとパターンの接続関係は論理関係、定義、帰属の関係、類似の関係、作用と結果の関係、推論の展開等、様々な関係を表現することが可能であり、また属性の継承、共通的特徴の継承と個々の特徴の識別ということも柔軟に表現することができる。
このように情報と情報の関係をパターンとパターンとの接続関係として設定していくことにより、情報は単独の状態で記録するのではなく、他の情報との関係をもった知識体系として記録することが可能となる。
情報と情報の関係は多岐の関係を有することになるが、情報間の関係(同じ意味、定義、反対の意味、類似、論理、原因、結果、詳細、概略、要約、関連情報等)をパターンの各処理フェーズ(思考プロセスの段階に対応)において適切なものを選択することにより、パターンからパターンの遷移を適切に制御することができる。
By converting the language that expresses human thought into patterns, and analyzing patterns and relationships between patterns, sentence types (normal sentences, question sentences, imperative sentences, etc.), features (good and evil, truth, truth, facts, definitions, Rules, common sense, explanations, hypotheses, predictions, opinions, impressions) and relationships (causes and consequences, events and reasons, explanations and conclusions, outlines and details, etc.) can be identified. Sentence types can be trained by machines so that humans can identify the types and characteristics of sentences. The sentence type can be identified as a normal sentence, a question sentence, a command sentence, a conditional sentence, etc. by analyzing words included in the sentence. Among the features of the sentence, human beings teach each machine about the good, evil, truth, truth, fact, definition, rule, and common sense, and let the machine learn. This can be done by identifying each information type (truth, truth, fact, definition, rule, common sense) and recording it with additional information when each information is input to the machine. Identification of other types of information (explanation, hypothesis, prediction, opinion, impression) is that the sentence type is not recorded as (truth, truth, fact, definition, rule, common sense) and the word ( I think, think, think, etc.) can be implemented by analyzing.
If the language is converted into a pattern and expressed as a set of patterns corresponding to the words that make up the sentence, and the semantic relationship between words is expressed by the connection relationship between the patterns, the meaning of the sentence is expressed between the patterns and the patterns. It can be expressed as a connection relationship. Input information includes word identification, word characteristics (part of speech, meaning), sentence element analysis, sentence structure analysis (subject, predicate, subject modification, predicate modification, modification relationship), sentence elements and Sentence element relationship analysis (same meaning, definition, opposite meaning, etc.) is performed, and the relationship between the sentence element and the sentence element is associated as a connection relationship between the pattern and the pattern, so that the relationship between the information and the meaning of the information is the pattern and the pattern The relationship and structure will be transformed. The relationship between patterns can be set not only for sentence elements and sentence elements but also for sentences and sentences, and sentences and sentences. This can be done by defining a group of patterns as a new pattern. When a characteristic word indicating a pattern-pattern relationship (cause and result, event and reason, explanation and conclusion, outline and details, etc.) is detected, the corresponding relationship between the corresponding patterns is recorded autonomously. This relationship between recorded patterns can be utilized when controlling transitions between patterns.
The connection relation between patterns can express various relations such as logical relation, definition, attribution relation, similar relation, relation between action and result, inference development, inheritance of attributes, common Feature inheritance and identification of individual features can also be expressed flexibly.
By setting the information-information relationship as a pattern-to-pattern connection relationship in this way, information is not recorded as a single state, but as a knowledge system that has relationships with other information. It becomes possible.
The relationship between information has various relationships, but the relationship between information (same meaning, definition, opposite meaning, similarity, logic, cause, result, detail, summary, summary, related information, etc.) By selecting an appropriate one in each processing phase (corresponding to the stage of the thought process), it is possible to appropriately control the pattern transition from pattern to pattern.

情報が入力されると情報に含まれている単語から意味を分析し、情報と等価な意味を示すパターンを生成する。このパターンは文の主語、述語、主語の修飾、述語の修飾、修飾関係を維持した状態で生成する。この生成したパターンを使って記録されているパターンと照合を行い、関連するパターンが存在するか否かの分析を行う。このように意味レベルで情報の検索を行うことができるので、文言が一致しなくても意味および内容的に関連のある情報を検索することが可能となる。所謂、連想記憶システムにおける検索を実施することができる。
新規に入力した情報は既に記録している知識体系と照合し、整合性および新規性について評価することができる。真理、真実、事実、定義、規則および常識として識別し記録している情報と整合しない場合は、その情報は誤っている可能性が高い。一方、整合しているか整合していないかの判断ができない場合は、その情報を判断できるだけの知識が未だ蓄えられていないと考えられるため、人間に通知し判断を求めることにする。人間がその情報は正しいと判断できる場合は、その結果を機械に通知することとする。機械はその情報を説明または仮説という位置づけで記録し、将来別の情報を得ることにより、真理、真実、事実、定義、規則および常識の組合せで論理展開し、整合性および妥当性を確認できた時に、評価を実施した信頼性の高い情報として識別し記録することとする。
このように、情報をパターンに変換し、情報の種類、特徴を識別、分類して記録し、情報と情報の関係(論理的な関係、類似関係、相反関係、類推等、様々な関係)をパターンとパターンの接続関係として表現することにより、情報を知識体系として構築することができる。
When information is input, a meaning is analyzed from words included in the information, and a pattern indicating a meaning equivalent to the information is generated. This pattern is generated while maintaining the sentence subject, predicate, subject modification, predicate modification, and modification relationship. The generated pattern is used for collation with the recorded pattern, and an analysis is performed as to whether or not there is a related pattern. Since information can be searched at the semantic level in this way, it is possible to search for information that is related in meaning and content even if the words do not match. A search in a so-called associative memory system can be performed.
The newly entered information can be checked against the already recorded knowledge system and evaluated for consistency and novelty. If it is inconsistent with the information identified and recorded as truth, truth, fact, definition, rule and common sense, the information is likely to be incorrect. On the other hand, if it is not possible to determine whether the information is consistent or not, it is considered that knowledge sufficient to determine the information has not yet been accumulated, so a human being is notified and a determination is made. If humans can judge that the information is correct, the result will be notified to the machine. The machine recorded the information as an explanation or hypothesis, and obtained another information in the future, and logically developed it with a combination of truth, truth, facts, definitions, rules and common sense, and was able to confirm consistency and validity Sometimes it will be identified and recorded as highly reliable information evaluated.
In this way, information is converted into patterns, information types and characteristics are identified, classified and recorded, and information-information relationships (logical relationships, similar relationships, reciprocal relationships, analogies, etc.) Information can be constructed as a knowledge system by expressing the connection relationship between patterns.

情報を知識体系として構築すると、知識体系を活用した問題解決が可能となる。まず、問題について知識体系を活用して分析し問題点の明確化および問題解決の目標を設定する。次に問題点を解決するための解決策案を抽出し適用する。適用後の状況を評価し、問題解決の目標に到達すれば処理としては完了になる。問題解決の目標に到達していなければ、再度新しい状態に対し、上記のプロセスを適用し、問題解決の目標に近づけていく。これらの処理についても自然言語で指示することにより、本機械はパターン間の接続関係に変換し処理を実行していく。状況に対応するパターンを分析することにより、問題が識別され、問題を解決する上で必要なパターンが逐次、起動し処理を実行する。処理により状況は変化し、変化した状況に対し更に同様のプロセスを作用させていくという方式で問題を解決していく。
問題の分析および問題点の明確化に関しても目標とする状態と現状の差異を検出、識別する方法を学習することにより実施する。また、各問題に対応する解決策も対応する処理を学習することにより実施する。学習はパターンの遷移の履歴を記録することにより行う。人間が教示した内容はパターンからパターンへの遷移として記録される。
学習が進むと、教示したプロセスはパターンからパターンへの遷移として自律的に実行され、教示した内容が実行されることになる。
人間の思考パターンの遷移は条件により遷移先が変わるという観点から、一般的に条件付き処理として表現することができる。本発明では言語の意味を解釈し、必要に応じて自律的に条件付処理に変換する。条件つき処理の条件の詳細については、該当する言語から検索用のパターンを生成し、自律的に検索する。検索した情報が条件を満足するか否かの判断を実施し、満足する場合は該当する処理の実行を行う。人間が知識として使用している問題解決策を、そのまま言語として入力しても、その意味を逐次、解釈し条件付処理を自律的に進めていき問題を解決していく。処理の途中で新規の情報が必要な場合は、その情報要求を通知し、該当する適切な情報が獲得されると、その内容に応じた処理を実施する。
学習フェーズにおいては、パターンが起動すると、そのパターンが起動する以前の設定した期間において起動したパターンの履歴が参照され、起動したパターンとの接続関係が強化される。実用フェーズにおいては、記録した起動履歴の一部のデータを使用して各記録ユニットに記録した起動の条件と合致するかを確認し、起動の条件と合致した場合に該当パターンを起動する。パターンが起動すると、起動パターンの履歴が更新され、新しい状態において上記の動作を繰り返し、起動の条件と合致したパターンを逐次、起動していく。
If information is constructed as a knowledge system, problem solving using the knowledge system becomes possible. First, the problem is analyzed using the knowledge system, the problem is clarified, and the goal of problem solving is set. Next, a solution plan for solving the problem is extracted and applied. The situation after application is evaluated, and if the goal of problem solving is reached, the process is completed. If the problem solving goal has not been reached, the above process is applied again to the new state to bring it closer to the problem solving goal. By instructing these processes in a natural language, this machine converts the connection relationship between patterns and executes the process. By analyzing the pattern corresponding to the situation, the problem is identified, and the pattern necessary for solving the problem is sequentially activated and the process is executed. The situation changes by the process, and the problem is solved by a system in which a similar process is further applied to the changed situation.
The problem analysis and problem clarification are also implemented by learning how to detect and identify the difference between the target state and the current situation. Moreover, the solution corresponding to each problem is implemented by learning the corresponding process. Learning is performed by recording a history of pattern transitions. The content taught by humans is recorded as a transition from pattern to pattern.
As learning progresses, the taught process is autonomously executed as a transition from pattern to pattern, and the taught content is executed.
The transition of human thought patterns can be generally expressed as a conditional process from the viewpoint that the transition destination changes depending on the condition. In the present invention, the meaning of the language is interpreted and autonomously converted into conditional processing as necessary. As for the details of the conditional processing conditions, a search pattern is generated from the corresponding language, and the search is performed autonomously. Judgment is made as to whether or not the retrieved information satisfies the condition, and if satisfied, the corresponding processing is executed. Even if a problem solution that humans use as knowledge is input as a language as it is, the meaning is sequentially interpreted and conditional processing proceeds autonomously to solve the problem. When new information is required during the process, the information request is notified, and when appropriate information is acquired, a process corresponding to the content is performed.
In the learning phase, when a pattern is activated, the history of the pattern activated in the set period before the pattern is activated is referred to, and the connection relationship with the activated pattern is strengthened. In the practical phase, a part of the recorded activation history data is used to check whether the activation condition recorded in each recording unit is met, and when the activation condition is met, the corresponding pattern is activated. When the pattern is activated, the history of the activation pattern is updated, the above operation is repeated in the new state, and the pattern that matches the activation condition is sequentially activated.

学習フェーズにおいては人間の指示により処理に対応する一連のパターンを逐次、起動していき、その履歴を記録することにより、指示された手順に従って、該当のパターンが起動していく。パターンの動作は静的なものではなく、プログラムの機能も有し動的な挙動を示す。パターン内に記録している内部パターンを使用して、必要な情報を検索し、処理を実施し、処理結果を必要な場所に格納する。また、パターンに記録している内部パターンを指定された配置に変換する等の処理を実施する。また、パターンの起動をトリガーとして様々な問題を解く処理プログラムを起動する。この様なダイナミックな挙動をするパターンを組み合わせることにより、数々の処理(数式処理、化学式処理、翻訳等)を実施させることができる。
本機械の全体動作はパターン制御器にて管理する。各パターンの遷移サイクルで、情報入力、情報分析(文の種類、構文、意味等)、情報評価(新規性、信頼性、妥当性、有用性等)、情報処理(問題・課題の解決策の生成、記録、情報出力等)を実施する。
本機械への教示はプログラミングすることなく、自然言語を逐次、入力していくことにより実施できる。(当然のことながらプログラミングにより教示しても良い)入力した言語情報は構文、意味、既に記録されている情報との関係が分析され、分析結果に応じて、対応するパターンが起動し、処理が実行される。入力情報の価値評価と記録、指示された命令の実行、問題・課題に対する解決策の生成等、数々の処理の実施が可能である。
プログラミングではプログラミング言語に従って、変数、定数の読出し、処理、処理結果の書込み、条件分岐による処理が実施される。本人工知能装置では入力した情報(言語)を分析・解釈し、分析・解釈結果に応じて、関連情報の検索、関連情報間の遷移(パターン間遷移)、条件に応じた処理の分岐、情報の処理(パターン処理)、生成した情報の記録(パターンの記録)を実施する。これらの処理により、本人工知能装置はプログラミング言語と同様に様々な情報の処理を行うことができる。(繰り返し処理等も実施できる)当然のことながら、通常のプログラミング処理を実施することも可能である。
本人工知能の特徴としては、通常のプログラミングによる処理だけではなく、プログラムではなく言語で入力した情報でも、分析・解釈することにより知識体系に蓄積したパターン(情報を検索し易い状態に変換したもの)を使用することにより、人間の思考パターンを模擬したパターン間の遷移を行い、問題を解決していくことができる。
本人工知能装置は人間の言語を理解するとともに、プログラムを実行することもできる。従って人間とインタフェースを取りながら、人間がこれまでに開発したプログラム群を動作させ、プログラム群の結果を総合して人間に報告させるという高度なシステムを構築することも可能である。
In the learning phase, a series of patterns corresponding to processing is sequentially activated by a human instruction, and by recording the history, the corresponding pattern is activated in accordance with the instructed procedure. The behavior of the pattern is not static but also has program functions and exhibits dynamic behavior. Using the internal pattern recorded in the pattern, the necessary information is searched, the processing is performed, and the processing result is stored in the necessary location. Also, processing such as conversion of the internal pattern recorded in the pattern into a designated arrangement is performed. In addition, a processing program that solves various problems is triggered by pattern activation. Numerous processes (formula processing, chemical formula processing, translation, etc.) can be performed by combining patterns having such dynamic behavior.
The overall operation of this machine is managed by the pattern controller. In each pattern transition cycle, information input, information analysis (sentence type, syntax, meaning, etc.), information evaluation (novelty, reliability, validity, usefulness, etc.), information processing (problem / problem solution) Generation, recording, information output, etc.).
Teaching to this machine can be performed by inputting natural language sequentially without programming. (Of course, it may be taught by programming) The input language information is analyzed for syntax, meaning, and relationship with already recorded information, and the corresponding pattern is activated and processed according to the analysis result. Executed. Numerous processes such as value evaluation and recording of input information, execution of instructed instructions, and generation of solutions to problems / issues are possible.
In programming, according to the programming language, reading of variables and constants, processing, writing of processing results, and processing by conditional branching are performed. This artificial intelligence device analyzes and interprets the input information (language), searches related information, transitions between related information (transition between patterns), branches processing according to conditions, information according to the analysis and interpretation results Process (pattern process) and recording of the generated information (pattern recording) is performed. Through these processes, the artificial intelligence apparatus can process various information in the same manner as a programming language. Naturally, it is also possible to carry out a normal programming process.
The feature of this artificial intelligence is not only the processing by normal programming, but also the information accumulated in the knowledge system by analyzing and interpreting the information input in the language instead of the program (the information converted into a state that is easy to search ) Can be used to make transitions between patterns simulating human thought patterns and solve problems.
This artificial intelligence device can understand human language and can also execute programs. Therefore, it is possible to construct an advanced system that operates a program group that has been developed so far while interfacing with a human and makes the result of the program group comprehensively reported to the human.

次に入力情報の一般化について説明する。
入力した文に含まれる単語の一部の特徴抽出および一般化を実施する。例えば、文中に出現する固有名詞は人物A、人物B、物C、物Dというように一般化する。逐次、文をパターンに変換し、変換したパターンを起動していく。この時、一般化したパターンも逐次、起動していくことになる。情報を数多く入力していくことにより、文を構成する特定のパターンとパターンの間の接続関係が強化されていく。入力した文は近傍の文と関係を有している。この関係は文を構成する単語と単語の関係で表現されるので、同じ関係は同一の単語間または類似の単語間の組合せで表現される場合が多い。文のパターンが起動する毎に、文に含まれる単語が起動するが、固有名詞等を一般化することにより、一般化した人物と人物または物との関係が強化されることになり、同様の文が出現する頻度が、固有名詞で表現した場合より多くなる。この効果により固有名詞に依存しない対象間の関係が抽出されることになる。また単語の意味、特徴等を起動すると、個々の単語間の関係では無く単語の特徴間の関係が強化され特徴面からの一般化が可能となる。この関係は特定のパターンとパターンとの接続関係が強化するため検出することができる。特に多くの文例から強化された関係は、一般性を有することになり、常識または一般的な考え方に対応するものが抽出されていく。この常識または一般的な考え方は入力する文章群に依存する。つまり、ある文化に対応する文章群を入力すると、その文化に対応した常識または一般的な考え方が抽出されることになる。同一文化での文章群を入力した場合は考え方が同等と考えられるため、同一文化での多数により強化された考え方が抽出されることになる。
同様に数々の問題とその解決策について一部の単語の特徴抽出および一般化を実施して動作させることにより、問題とその解決策について一般化した関係を抽出することができる。情報入力による学習が進行すると本機械は類似の問題に対して解決方法を自律的に生成することができるようになる。
Next, generalization of input information will be described.
Feature extraction and generalization of a part of a word included in an input sentence For example, proper nouns appearing in sentences are generalized as person A, person B, object C, object D. Sequentially, sentences are converted into patterns, and the converted patterns are activated. At this time, generalized patterns are also activated sequentially. By inputting a lot of information, the connection between specific patterns constituting a sentence is strengthened. The input sentence has a relationship with a nearby sentence. Since this relationship is expressed by the relationship between words constituting a sentence, the same relationship is often expressed by a combination between the same words or similar words. Every time a sentence pattern is activated, a word included in the sentence is activated. By generalizing proper nouns, etc., the relationship between generalized persons and persons or objects will be strengthened. Sentences appear more frequently than when expressed with proper nouns. With this effect, the relationship between objects that does not depend on proper nouns is extracted. In addition, when the meaning, features, etc. of the word are activated, the relationship between the features of the word rather than the relationship between the individual words is strengthened, and generalization from the feature side becomes possible. This relationship can be detected because the connection relationship between specific patterns is strengthened. In particular, relationships strengthened from many sentence examples have generality, and those corresponding to common sense or general ideas are extracted. This common sense or general idea depends on the text group to be entered. In other words, when a sentence group corresponding to a certain culture is input, common sense or a general idea corresponding to that culture is extracted. When a group of sentences in the same culture is input, the way of thinking is considered to be equivalent. Therefore, the way of thinking strengthened by many people in the same culture is extracted.
Similarly, by extracting and generalizing the characteristics of some words for a number of problems and their solutions and operating them, a generalized relationship between the problems and their solutions can be extracted. As learning by information input progresses, this machine can autonomously generate a solution to a similar problem.

以下では情報を条件部と処理部に識別し文構造として整理した状態でパターン記録器に記録する動作について説明する。情報の単語を分析することにより情報の条件部と処理部を識別することができる。例えば、「A」「が」「B」「の時」「C」「を実施せよ」という文では[「A」「が」「B」「の時」]が条件部であり、[「C」「を実施せよ」]は処理部である。また、「A」「が」「B」「なら」「C」「は」「D」「である」という文では[「A」「が」「B」「なら」]が条件部であり、[「C」「は」「D」「である」]は処理部である。
このように文を条件部と処理部に分け、条件付処理のパターン間接続を有した構造に変換する。また、条件部のパターンが起動すると、自律的に条件部が成立しているか否かを確認する処理を起動するようにする。このように文を条件付き処理の構造を有したパターンに変換すると、条件付処理で表現できる文が逐次、条件の成立性を確認しながら処理を進めていくという能動的な思考動作を実現することができる。一般的に人間の問題解決、行動決定は条件付処理で表現することができる。人間の問題解決策および行動決定策を自然言語で入力することにより本機械は自律的に条件付処理に変換した上で、条件の成立性を確認しながら処理を進めていくという能動的な思考動作が可能となる。人間の問題解決および行動決定に相当する動作をプログラミングすることなく自然言語で表現された知識(問題解決および行動決定に関する手順および思考方法)を入力することにより、人間が思考により問題解決または行動決定するように問題解決または行動決定を自律的に実施することが可能となる。
In the following, description will be given of the operation of recording information in the pattern recorder in a state in which information is identified as a conditional part and a processing part and organized as a sentence structure. By analyzing the information word, the condition part and the processing part of the information can be identified. For example, in the statements “A”, “GA”, “B”, “NO”, “C”, “do it”, [“A”, “GA”, “B”, “NO” ”is the condition part, and [“ C "Perform""is a processing unit. Also, in the sentence “A” “ga” “B” “if” “C” “ha” “D” “is” [“A” “ga” “B” “if”] is the condition part, [“C” “is” “D” “is”] is a processing unit.
In this way, the sentence is divided into a conditional part and a processing part, and converted into a structure having inter-pattern connection for conditional processing. When the condition part pattern is activated, a process for autonomously confirming whether or not the condition part is established is activated. When a sentence is converted into a pattern with a conditional processing structure in this way, an active thinking action is realized in which sentences that can be expressed by conditional processing proceed sequentially while confirming the validity of the condition. be able to. In general, human problem solving and action determination can be expressed by conditional processing. By inputting human problem-solving and action-determining measures in natural language, this machine autonomously converts it into conditional processing, and then proceeds with the processing while confirming the validity of the conditions. Operation is possible. By inputting knowledge (procedures and thinking methods related to problem solving and action determination) expressed in natural language without programming the actions corresponding to human problem solving and action determination, human beings can solve problems or make action decisions through thinking. As a result, problem solving or action determination can be carried out autonomously.

図1はこの発明の一実施例における人工知能装置の構成を示した図である。図1において1は情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器である。2はパターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器である。3は入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器である。4は人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器である。5はパターンを情報に変換するパターン逆変換器である。6はパターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器である。  FIG. 1 is a diagram showing a configuration of an artificial intelligence apparatus according to an embodiment of the present invention. In FIG. 1, reference numeral 1 denotes a pattern converter that converts information (word, language, sentence, knowledge, mathematical formula, symbol, image, voice) into a pattern. Reference numeral 2 denotes a pattern recorder for recording patterns, patterns and conditions for starting a processing program, connection relations between patterns, relations with patterns having connection relations, and pattern analysis results. 3 is the source of information, reliability, novelty, field, theme, topic, intention, subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, and how , Why, why, the type of sentence (ordinary sentence, question sentence, imperative sentence, conditional sentence, truth, fact, rule, common sense, right and wrong, definition, logic, explanation, hypothesis, prediction, opinion, impression, rumors), Pattern analysis that analyzes formulas, physical formulas, chemical formulas, symbols, and relationships with recorded related information, organizes them into structures that can be searched using words, sentence elements, sentences, and analysis results, and records them in a pattern recorder It is a vessel. Reference numeral 4 denotes a pattern controller that performs pattern registration, activation, change, processing, control, and activation of a processing program in accordance with human instructions, pattern states, and pattern analysis results. Reference numeral 5 denotes a pattern inverse converter for converting a pattern into information. Reference numeral 6 denotes a pattern irradiator that records a pattern activation history, analyzes the situation, context, and problem using the recorded activation history information, and activates a pattern and a processing program necessary for problem solving.

次に動作について説明する。
情報(言語)を語列の形で入力すると、各語を検出する記録モジュールが励起する。各単語に対応する記録モジュールの接続情報記録部には単語を識別するための語列の型が記録されており、単語の語列と合致する語列の履歴が照射されると該当する単語の記録モジュールが励起する。各単語の記録モジュールは単語の意味、品詞、対応する外国語等を記録した記録モジュールと接続されており、入力した文の構文分析に使用される。構文分析された文は主語、主語の修飾、述語、述語の修飾(何を、いつ、どこで、なぜ、どのように)を識別し、検索し易い形に整理してパターン記録器に記録する。パターンは文の構造を維持した形で記録するので意味的な検索(いつ、どこで、誰が、何を、何故、どのように)を容易に実施できる。
図2は語列の励起履歴の例を示したものである。本例で入力した文は「わたしはせんせいです。」(私は先生です。)である。語の励起履歴が各単語の接続情報記録部に照射され照合が行われる。単語に対応する語列および一致した語数を検出すると該当の単語が励起する。単語が検出されると検出した単語の語列を除いた励起履歴を使って次の単語の検出を行う。この励起履歴により次の単語が検出できない場合には、前の単語の検出は誤っていたと判断し、別の単語候補に対して整合性を分析し単語の候補を絞り込む。
単語、数、数式、記号からパターンへの変換は識別できるものであれば特に形式に制約は無い。またパターンの組合せも、パターンである。本人工知能のプロトタイプでは各単語に識別番号を割り当てることにより実施した。(例えば「私」という単語の単語識別番号は「0」、助詞「は」の単語識別番号は「1」、「先生」「2」、・・・「人工知能」「169」、「職業」「170」・・・等々)新規の単語が発生した場合には識別番号を逐次追加することにより対応することができる。数々の思考・概念はパターンの組合せ(本例では各単語、記号に対応付けした識別番号の組合せを主語、述語、修飾関係を識別した構造に整理している)で表現することができる。
図3は語列から単語、単語の品詞、意味を分析し、さらに文要素、文の構造を分析する機能図を示したものである。語列から単語を識別し、単語および単語と同期して励起する品詞および意味から文要素を識別していく。文要素を識別することにより主語、主語の修飾、述語、述語の修飾(何を、いつ、どこで、なぜ、どのように)を識別し、整理してパターン記録器に記録する。
このように識別、整理して文を記録すると非常に有益である。情報を入力した際に既に記録している情報との照合を文要素毎に実施できるので新規性の検出および記録している情報との差異を正確かつ容易に分析することができる。
図4は文の構造の分析例を示したものである。語列から単語、単語識別番号、品詞、品詞識別番号、文要素が識別されている。
図5は記録した情報から質問に対する回答を生成する機能図を示したものである。
すると、残りのパターンと組み合わせることにより質問に対応する検索パターンを生成することができる。質問の回答として期待する文要素が何なのか(主語、主語の修飾部、述
ンをパターン記録器に照射し、相関があるパターンを検索する。
あれば回答の候補となる。
図6は情報をパターン(本例では識別番号)に変換しパターン記録器に記録している様子を示したものである。入力情報の単語、記号、数字、式はパターン(識別番号)に変換し、主語、主語の修飾、述語、述語の修飾(何を、いつ、どこで、なぜ、どのように)を検索し易い形に整理して記録している。図7は入力した情報と記録している情報との比較動作例である。情報は文要素に整理されているので文要素毎の比較が可能である。また、比較も完全一致、部分的一致、類似、反対の意味の検出等、様々な比較が可能である。記録モジュールに記録している単語間には同じ意味、類似、反対の意味、関連等の関係を定義することができる。入力した情報に含まれる単語から、これらの関連性を有する単語を使用した照合および比較を実施することにより記録している情報との様々な関係を抽出することができる。
Next, the operation will be described.
When information (language) is input in the form of a word string, a recording module that detects each word is excited. In the connection information recording unit of the recording module corresponding to each word, a word string type for identifying the word is recorded, and when the history of the word string matching the word word string is irradiated, The recording module is excited. Each word recording module is connected to a recording module that records word meaning, part of speech, and corresponding foreign language, and is used for syntactic analysis of the input sentence. The syntax-analyzed sentence identifies the subject, subject modification, predicate, and predicate modification (what, when, where, why, and how), organizes them in a form that is easy to search, and records them in a pattern recorder. Patterns are recorded in a way that preserves the structure of the sentence, so semantic searches (when, where, who, what, why, how) can be easily performed.
FIG. 2 shows an example of the excitation history of the word string. The sentence entered in this example is "I am a teacher" (I am a teacher). The word excitation history is irradiated to the connection information recording unit of each word and collation is performed. When the word string corresponding to the word and the number of matched words are detected, the corresponding word is excited. When a word is detected, the next word is detected using an excitation history excluding the detected word string. If the next word cannot be detected by this excitation history, it is determined that the previous word has been detected incorrectly, and the consistency of another word candidate is analyzed to narrow down the word candidates.
There are no restrictions on the format of words, numbers, mathematical expressions, and conversion from symbols to patterns as long as they can be identified. A combination of patterns is also a pattern. In this artificial intelligence prototype, each word was assigned an identification number. (For example, the word identification number of the word “I” is “0”, the particle identification number of the particle “ha” is “1”, “teacher” “2”,... “Artificial intelligence” “169”, “profession” “170”, etc.) When a new word is generated, it can be dealt with by sequentially adding an identification number. Numerous thoughts / concepts can be expressed by combinations of patterns (in this example, combinations of identification numbers associated with words and symbols are arranged in a structure in which the subject, predicate, and modification relationship are identified).
FIG. 3 is a functional diagram for analyzing words, parts of speech and meanings of words from a word string, and further analyzing sentence elements and sentence structures. The word is identified from the word string, and the sentence element is identified from the part of speech and the meaning excited in synchronization with the word and the word. By identifying sentence elements, the subject, subject modification, predicate, and predicate modification (what, when, where, why, how) are identified, organized, and recorded in a pattern recorder.
It is very useful to record sentences by identifying and organizing in this way. Since the collation with the information already recorded when the information is input can be performed for each sentence element, the novelty can be detected and the difference from the recorded information can be analyzed accurately and easily.
FIG. 4 shows an example of sentence structure analysis. A word, a word identification number, a part of speech, a part of speech identification number, and a sentence element are identified from the word string.
FIG. 5 is a functional diagram for generating an answer to a question from recorded information.
Then, a search pattern corresponding to the question can be generated by combining with the remaining patterns. What are the expected sentence elements to answer questions (subjects, subject modifiers, descriptions)
The pattern recorder is irradiated to search for a correlated pattern.
If there is a candidate for the answer.
FIG. 6 shows how information is converted into a pattern (identification number in this example) and recorded in a pattern recorder. Words, symbols, numbers, and expressions in the input information are converted into patterns (identification numbers), and the subject, subject modification, predicate, predicate modification (what, when, where, why, how) can be easily searched It is organized and recorded. FIG. 7 shows an example of a comparison operation between input information and recorded information. Since information is organized into sentence elements, it is possible to compare each sentence element. In addition, various comparisons such as complete match, partial match, similarity, and detection of opposite meanings are possible. Relationships such as the same meaning, similarities, opposite meanings, and relationships can be defined between words recorded in the recording module. Various relations with the recorded information can be extracted from the words included in the input information by performing collation and comparison using these related words.

入力した情報は次のように評価して処置を決める。
第1段階では情報の信頼性評価を実施する。これは情報源(いつ、誰、何処からの情報か?)の信頼性を確認することにより実施可能である。
第2段階では情報の分野/テーマ分析を実施する。これは情報に含まれている単語から分野/テーマについて検出することが可能である。
第3段階では情報の種類識別を実施する。入力情報の種類の内、真理、真実、事実、定義、規則、常識、善悪については人間からの指定に従うこととする。説明、仮説、予測、意見、感想等については情報に含まれる単語(だろう、思う、考える・・等)から識別することが可能である。平常文、疑問文、命令文、条件文、感嘆文の識別についても含まれる単語から実施することが可能である。
第4段階では関心度評価を実施する。関心ある情報の分野/テーマを事前に設定しておき、情報の分野/テーマ分析結果と照合し合致するか否かを確認することにより評価することが可能である。
第5段階では新規性評価を実施する。これは入力情報および入力情報と意味的に等価なパターンを記録領域に照射し、関連するパターンの有無を確認する。関連するパターンが検出された場合はパターン間の相違点について比較し、新規パターンが有るか否かを評価することにより実施可能である。
第6段階では妥当性評価を実施する。これは入力情報と関連情報のパターンを単語間関係について定義した記録領域に照射し、文要素毎に整合、不整合を評価する。不整合が検出された場合は、入力情報と関連情報の信頼性(真理、真実、事実、定義、規則、常識、善悪、情報源の信頼度)を評価し、信頼性の高いものを優先的に採用し記録することとする。また、規則、常識、道徳等に反する情報は人工知能装置の知識体系に組み込まず棄却する。
図8は入力した情報を評価して処置(記録、破棄)を実施する機能図である。
The input information is evaluated as follows to determine the treatment.
In the first stage, information reliability is evaluated. This can be done by checking the reliability of the information source (when, who, where from?).
In the second stage, an information field / theme analysis is performed. It is possible to detect the field / theme from the words included in the information.
In the third stage, information type identification is performed. Among the types of input information, truth, truth, facts, definitions, rules, common sense, good and evil shall be specified by humans. Explanations, hypotheses, predictions, opinions, impressions, etc. can be identified from the words (I think, think, think ...) included in the information. Identification of ordinary sentences, question sentences, command sentences, conditional sentences, and exclamation sentences can also be performed from included words.
In the fourth stage, the interest level is evaluated. It is possible to evaluate by setting a field / theme of the information of interest in advance and checking whether or not it matches with the field / theme analysis result of the information.
In the fifth stage, novelty evaluation is conducted. This irradiates the recording area with input information and a pattern that is semantically equivalent to the input information, and confirms the presence or absence of a related pattern. When a related pattern is detected, it can be implemented by comparing the differences between the patterns and evaluating whether or not there is a new pattern.
In the sixth stage, validity evaluation is performed. This irradiates the recording area where the relation between words is defined with the pattern of the input information and the related information, and evaluates the matching and mismatching for each sentence element. If inconsistencies are detected, the reliability of input information and related information (truth, truth, fact, definition, rules, common sense, good and evil, reliability of information source) is evaluated, and the one with high reliability is given priority. Will be adopted and recorded. Information that violates rules, common sense, morals, etc. is rejected without being incorporated into the knowledge system of the artificial intelligence device.
FIG. 8 is a functional diagram for evaluating the input information and performing a treatment (recording and discarding).

図9は励起したパターンの履歴がパターン照射器に記録される動作について示している。ある記録ユニットが励起すると、それ以前に励起したパターンの履歴のデータから当該パターンとの接続関係のデータを生成して当該パターンの記録ユニットの接続関係記録部に記録し、励起に関連するパターンとの接続関係が強化する動作について示している。パターン照射器は現時点から設定した過去までのパターンの履歴のデータと各記録ユニットの接続関係記録部に記録した接続関係のデータとを照合させ、相関が大きい記録ユニットを励起する。励起のパターンが繰り返し出現すると、繰り返し出現するパターンとの接続関係が特に強化される。この動作により励起したパターンとパターンの接続関係が強化され学習が進んでいく。学習データを逐次、入力していくと各学習データに対応するパターンとパターンとの接続関係が強化していき学習データ間の接続関係が知識体系の中に生成されていく。学習が進んだ段階で一部の学習データを入力すると知識体系に生成された学習データに対応するパターン間の接続関係に沿ってパターンが遷移し、学習した内容が想起されていく。つまり、あるテーマに関するキーワードを入力することにより学習した関連情報が逐次、検索することが可能となる。学習データの遷移先が複数の場合は、何を知りたいのかという目標に応じてパターン間の接続関係の関係性を選択することにより、複数の関連情報の中から欲しい情報を絞っていくことができる。
本人工知能は人間の思考または情報をパターンとして表現し、パターンからパターンへ自律的に遷移させることが可能である。この自律的な遷移を実現する記録ユニットの機能概要を図10に示す。記録ユニットは情報に関する各種の内容を記録している。記録する内容の例としては、情報の内容を表現するパターン、情報間の関係を表現する他の記録ユニットとの接続情報(接続数、接続先、接続先との関係性)、記録ユニットが励起するための条件(照合条件、合致条件)、パターンの分析結果(情報の源泉、信頼性、新規性、分野、テーマ、文の種類)
等がある。文の種類(平常文、疑問文、命令文、条件文、感嘆文、真実、事実、規則、常識、定義、論理、説明、仮説、予測、意見、感想、噂)、相対的意味を有する単語の絶対化、数式、化学式、記号等を識別し分析結果として記録する。また、情報の内容であるパターンは主語、主語の修飾、述語、述語の修飾、修飾関係、文の種類、数式、物理式、化学式、記号、いつ、どこで、誰が、何を、どのように、なぜ、したのか、および記録している関連情報との関係を分析し、情報を検索し易い構造に整理して記録する。情報が入力されると情報の分野、テーマ、文の種類、構文(主語、主語の修飾、述語、述語の修飾、修飾関係、文の種類、数式、物理式、化学式、記号、いつ、どこで、誰が、何を、どのように、なぜ、したのか)を分析し、記録している情報との照合を行い情報の価値(信頼性、新規性、妥当性、整合性、善悪、既知の情報との関係)を評価する。評価した結果、安全で価値のある情報は分野、テーマ、既知の情報との関係を整理して記録器に記録する。
情報間の関係としては原因と結果、事象と理由、説明と結論、概略と詳細、類似の意味、反対の意見、本文と要約、本文と詳細情報等、様々なものがある。これらの関係性の設定は人間の指示または情報と情報の関係を学習することによる自律設定により実施する。
このように情報は関連情報と関係性を有した知識体系として記録されるので、ある情報が入力され、該当する記録ユニットが励起すると、励起した記録ユニットから関連する記録ユニットの情報(接続先、接続先との関係性)を呼び出し、その中から適切な記録ユニットを逐次、励起し必要な情報を検索していくことが可能となる。記録ユニットの励起をモニターすることにより人工知能装置における処理の中心が現在どこにあるかを確認することができる。人間が思考する場合において、思考または意識の中心が現在何かを確認することに相当する。また、従来型のプログラムで例えればコンパイルした命令コードの実行ポインタの位置をプログラム実行中に逐次、表示することに相当する。
FIG. 9 shows an operation in which the history of the excited pattern is recorded in the pattern irradiator. When a certain recording unit is excited, it generates connection relation data from the history data of the pattern excited before that and records it in the connection relation recording section of the recording unit of the pattern. It shows the operation that strengthens the connection relationship. The pattern irradiator collates the history data of the pattern set from the present time to the past with the connection relation data recorded in the connection relation recording section of each recording unit, and excites the recording unit having a large correlation. When the excitation pattern appears repeatedly, the connection relationship with the repeatedly appearing pattern is particularly strengthened. By this operation, the connection relationship between the excited pattern and the pattern is strengthened and learning proceeds. As learning data is sequentially input, the connection relationship between patterns corresponding to each learning data is strengthened, and a connection relationship between learning data is generated in the knowledge system. When a part of the learning data is input at the stage of learning, the pattern transitions along the connection relation between the patterns corresponding to the learning data generated in the knowledge system, and the learned content is recalled. That is, it becomes possible to sequentially search related information learned by inputting a keyword related to a certain theme. When there are multiple transition destinations of learning data, it is possible to narrow down the desired information from multiple related information by selecting the connection relationship between patterns according to the goal of what you want to know it can.
This artificial intelligence expresses human thought or information as a pattern and can autonomously transition from pattern to pattern. FIG. 10 shows a functional outline of the recording unit that realizes this autonomous transition. The recording unit records various contents related to information. Examples of content to be recorded include patterns that express information content, connection information with other recording units that express the relationship between information (number of connections, connection destinations, relationships with connection destinations), and excitation of recording units Conditions (matching conditions, matching conditions), pattern analysis results (information source, reliability, novelty, field, theme, sentence type)
Etc. Sentence type (ordinary sentence, question sentence, imperative sentence, conditional sentence, exclamation sentence, truth, fact, rule, common sense, definition, logic, explanation, hypothesis, prediction, opinion, impression, rumors), word with relative meaning Are identified and recorded as analysis results. In addition, the pattern that is the content of the information is the subject, subject modification, predicate, predicate modification, modification relationship, sentence type, formula, physical formula, chemical formula, symbol, when, where, who, what, how, Analyze why and why, and the relationship with the recorded related information, and organize and record the information in a structure that is easy to search. When information is entered, the field of information, theme, sentence type, syntax (subject, subject modification, predicate, predicate modification, modification relationship, sentence type, formula, physical formula, chemical formula, symbol, when, where, Analyze who, what, how and why) and compare it with the recorded information to determine the value of the information (reliability, novelty, validity, integrity, good / bad, known information) The relationship). As a result of the evaluation, safe and valuable information is recorded in a recorder by organizing the relationship with the field, theme, and known information.
There are various relationships between information such as causes and results, events and reasons, explanations and conclusions, outlines and details, similar meanings, opposite opinions, texts and summaries, texts and details. Setting of these relationships is performed by an autonomous setting by learning a human instruction or a relationship between information and information.
Since information is recorded as a knowledge system having a relationship with related information in this way, when certain information is input and the corresponding recording unit is excited, information on the related recording unit (connection destination, (Relationship with connection destination) can be called, and the appropriate recording unit can be sequentially excited to search for necessary information. By monitoring the excitation of the recording unit, it is possible to confirm where the processing center in the artificial intelligence apparatus is currently located. In human thinking, this is equivalent to confirming what is currently the center of thought or consciousness. For example, in the case of a conventional program, this corresponds to sequentially displaying the position of the execution pointer of the compiled instruction code during program execution.

本人工知能では入力情報を既に評価済みの知識体系と妥当性、整合性を評価した後に知識体系に組み入れていくので、人間が教育を受けながら正しく知能を向上させていくのと同様に誤った情報を排除しつつ知識体系を構築していくことができる。構築している知識体系には単語間の意味関係(同じ意味、反対の意味等)についても識別することができるので、誤った情報、規則、道徳に反する情報は仮に入力されても内部に構築した知識体系と整合性が取れないということを検出し、知識体系に組み込まずに廃棄するので安全である。
このように情報を知識体系として構築すると、知識体系を活用した問題解決が可能となる。図11は問題を検出して自律的に解決する方法の機能図について示したものである。状態から問題を検出し、検出した問題に対応する解決策を適用していくことにより段階的に問題を解決していく。図12は本動作についてパターンの遷移で表現したものである。
まず、問題について知識体系を活用して分析し問題点の明確化および問題解決の目標を設定する。次に問題点を解決するための解決策を抽出し適用する。適用後の状況を評価し、問題解決の目標に到達すれば処理としては完了になる。問題解決の目標に到達していなければ、再度新しい状態に対し、上記のプロセスを適用し、問題解決の目標に近づけていく。
問題の分析および問題点の明確化に関しても目標とする状態と現状の差異を検出、識別する方法を学習することにより実施する。また、各問題に対応する解決策も対応する処理を学習することにより実施する。学習は人間の指示による設定またはパターンの励起の履歴を記録することにより行う。人間が教示した内容はパターンからパターンへの励起の遷移として記録される。学習が進むと、教示したプロセスはパターンからパターンへの遷移として自律的に実行され、教示した内容が実行されることになる。
問題を自律的に検出し問題を解決するための処理プログラムを自律的に起動することができると高度な会話システムを構成することができる。話者(相手)の会話が入力すると入力文について分析を行い文の種類、話者の関心、意図を分析する。分析結果に応じた処理プログラムが起動し会話を成立させるために必要な処理を実施する。処理プログラムの例としては挨拶、相手の情報収集、相手の会話の評価、会話のための情報検索、問題および質問への回答生成、問題を解くプログラム、相手への質問、確認等がある。また情報検索は会話の単語を検索キーとして連想記憶的な情報抽出、情報間の関係性を使用した関連情報の検索を実施し、様々な会話の状況への対応が可能である。会話により話者および関心のあるテーマに関して新規で有用な情報が収集されるとパターン記録器に記録し知識の拡大を行っていくことが可能である。
In this artificial intelligence, the input information is incorporated into the knowledge system after evaluating the validity and consistency with the already evaluated knowledge system, so it is erroneous as if human beings are improving their intelligence correctly while receiving education. It is possible to build a knowledge system while excluding information. The built knowledge system can also identify semantic relationships between words (same meanings, opposite meanings, etc.), so if wrong information, rules, or information that violates morals is entered, it is built internally It is safe because it detects that it is not consistent with the knowledge system and discards it without incorporating it into the knowledge system.
If information is constructed as a knowledge system in this way, problem solving using the knowledge system becomes possible. FIG. 11 is a functional diagram of a method for detecting and autonomously solving a problem. The problem is solved step by step by detecting the problem from the state and applying a solution corresponding to the detected problem. FIG. 12 represents this operation by pattern transition.
First, the problem is analyzed using the knowledge system, the problem is clarified, and the goal of problem solving is set. Next, a solution for solving the problem is extracted and applied. The situation after application is evaluated, and if the goal of problem solving is reached, the process is completed. If the problem solving goal has not been reached, the above process is applied again to the new state to bring it closer to the problem solving goal.
The problem analysis and problem clarification are also implemented by learning how to detect and identify the difference between the target state and the current situation. Moreover, the solution corresponding to each problem is implemented by learning the corresponding process. Learning is performed by recording a setting by a human instruction or a history of pattern excitation. The content taught by humans is recorded as an excitation transition from pattern to pattern. As learning progresses, the taught process is autonomously executed as a transition from pattern to pattern, and the taught content is executed.
If a processing program for autonomously detecting a problem and solving the problem can be activated autonomously, an advanced conversation system can be configured. When the conversation of the speaker (the other party) is input, the input sentence is analyzed, and the kind of sentence, the interest of the speaker, and the intention are analyzed. A processing program corresponding to the analysis result is activated and performs processing necessary to establish a conversation. Examples of processing programs include greetings, collecting information about the other party, evaluating the other party's conversation, searching for information for the conversation, generating answers to problems and questions, solving the problem, questions to the other party, and confirmation. In addition, information retrieval can be performed for various conversation situations by extracting associative memory information using a conversation word as a search key and retrieving related information using the relationship between information. When new and useful information about the speaker and the topic of interest is collected through conversation, it can be recorded in a pattern recorder and expanded.

本人工知能の動作例を次に示す。人間から与えられた情報を使って会話および数々の問題を解くことができることを説明する。本問題に回答するために必要な情報(源泉)については事前に人工知能に教示した。次の会話では人工知能が知識を活用しながら人間との会話、質問に対応している。各会話において人工知能は特徴的なパターンを検出し問題の認識、識別、問題を解くために必要な処理プログラムの起動を自律的に行い、新しく生成した情報は知識としてパターン記録器に記録し知識の拡大を行っている。各会話における人工知能の処理の詳細については(処理1)〜(処理18)に示す。
NO(会話番号)人間と人工知能の会話
1 人間:初めまして
2 人工知能:私はアトムです(処理1)
3 私は人工知能です
4 人間:私は太郎です
5 人工知能:私の職業はアシスタントです(処理2)
6 太郎さんの職業は何ですか
7 人間:私の職業は教師です
8 人工知能:教師は大変ですね(処理3)
9 人間:私は物理を授業で教えています
10 人間:私はテニスを放課後部活で指導します
11 人工知能:太郎さんの趣味は何ですか(処理4)
12 人間:私の趣味はスポーツです
13 人工知能:スポーツと言えば(処理5)
14 ケイ選手は昨日テニス大会で優勝しました
15 ケイ選手は決勝戦でストレート勝ちです
16 すごいですね
17 人間:職業が何故アシスタントですか
18 人工知能:皆さんの役に立ちたいからです(処理6)
19 人間:アシスタントについて詳しく教えてください
20 人工知能:私は問題解決を支援します(処理7)
21 私は生活上の問題を扱えます
22 私は物理の問題を扱えます
23 私は数学の問題を扱えます
24 私は相談相手になります
25 人間:それでは質問に答えてください
26 人工知能:了解しました(処理8)
27 人間:生活上の問題です
28 人間:レストランについて教えてください
29 人間:どの店の予算が8000円以下ですか
30 人工知能:店Aの予算店Bの予算店Cの予算店Eの予算が8000円以下です (処理9)
31 人間:更にどの店の距離が1km以下ですか
32 人工知能:店Aの距離店Cの距離店Eの距離が1km以下です(処理10)
33 人間:更にどの店のおすすめが日本酒ですか
34 人工知能:店Cのおすすめ店Eのおすすめが日本酒です(処理11)
35 人間:生活上の問題です
36 人間:リンゴ3個の値段は300円です
37 人間:リンゴ5個の値段は何ですか
38 人工知能:リンゴ5値段は500円です(処理12)
39 人間:数学の問題です
40 人間:x0は4です
41 人間:x1は5です
42 人間:[2 d2x/_dtdt+3 d_x/_dt+1=0]
43 人工知能:方程式の解は以下です(処理13)
x=A exp(p1t)+B exp(p2t)
p1=−1 p2=−0.5
A=−14 B=18
44 人間:数学の問題です
45 人間:[f=x**2+2x−4]
46 人間:[g=2x]
47 人間:fとgの交点は何ですか
48 人工知能:fとgの交点は−2と2です(処理14)
49 人間:fとgの囲む面積は何ですか
50 人工知能:fとgの囲む面積は10.6667です(処理15)
51 人間:物理の問題です
52 人間:ボールの初速はvです
53 人間:ボールの迎角はtです
54 人間:vは10m/sです
55 人間:tは45degです
56 人間:ボールの飛距離は何ですか
57 人工知能:飛距離は 10.2041 mです(処理16)
58 人間:物理の問題です
59 人間:バネの力は−kxです
60 人間:ダンパーの力は−d dx/dtです
61 人間:重力の力は+m_gです
62 人間:バネの力が物体Cに働きます
63 人間:ダンパーの力が物体Cに働きます
64 人間:重力の力が物体Cに働きます
65 人間:何が物体Cに働きますか
66 人工知能:バネの力がダンパーの力が重力の力が働きます(処理17)
67 人間:gは9.8です
68 人間:m_は1です
69 人間:k は1です
70 人間:d は1です
71 人間:運動の方程式は何ですか
72 人工知能:運動方程式は以下です(処理18)
g =9.8
m_=1
k =1
d =1
d2x/_dtdt+(d/m_)d_x/_dt+(k/m_)x=g
★★★★★以下は人工知能の処理(詳細)★★★★★
(処理1)………………………………
(1) 関連情報の抽出
挨拶文(はじめまして)検出
話者(相手)の情報(名前、職業)収集プログラム起動
話者(相手)の情報(名前)は未記録
(2) 人工知能(出力)
私はアトムです
私は人工知能です
(処理2)………………………………
(1) 関連情報の抽出
話者(相手)の情報(名前)を検出
話者(相手)の情報(名前)は 太郎
話者(相手)の情報(名前)は既に記録済
話者(相手)の情報(職業)は未記録
(2) 人工知能(出力)
私の職業はアシスタントです
太郎さんの職業は何ですか
(処理3)………………………………
(1) 関連情報の抽出
話者(相手)の情報(名前)は既に記録済
話者(相手)の情報(名前)は 太郎
職業に関する回答を検出
話者(相手)の情報(職業)は 教師
話者(相手)の職業に対する感想を検索・出力
(2) 人工知能(出力)
教師は大変ですね
(処理4)………………………………
(1) 関連情報の抽出
話者(相手)の情報(趣味)収集プログラム起動
話者(相手)の情報(趣味)は未記録
(2) 人工知能(出力)
太郎さんの趣味は何ですか
(処理5)………………………………
(1) 関連情報の抽出
話者(相手)の情報(趣味)を検出
話者(相手)の情報(趣味)は スポーツ
話者(相手)の趣味に対する関連ニュースを検索・出力
検出した関連情報の分析(「ストレート勝ち」を検出
検出した関連情報の評価(「ストレート勝ち」に対する評価「すごいですね」を検索
検出した関連情報の評価、感想を出力
(2) 人工知能(出力)
スポーツと言えば
ケイ選手は昨日テニス大会で優勝しました
ケイ選手は決勝戦でストレート勝ちです
すごいですね
(処理6)………………………………
(1) 関連情報の抽出
qqtype=6;//質問の種類を(何故**ですか)に設定
接続関係有り:接続数= 6
senconrel[i][j]= 1
senconrel[i][j]= 1
senconrel[i][j]= 1
senconrel[i][j]= 1
senconrel[i][j]= 1
senconrel[i][j]= 5
接続関係に理由の項目有り:検出
接続関係(理由)の接続先の文番号= 28
接続関係(理由)の文出力:皆さんの役に立ちたいからです
(2) 人工知能(出力)
皆さんの役に立ちたいからです
(処理7)………………………………
★★文間の接続関係を利用した関連情報のアクセス(例)★★
基準の文の表示
私の職業はアシスタントです
基準の文と関連文(接続関係)の数は 6
1 私は問題解決を支援します
(基準文との関係は[詳細])
2 私は生活上の扱えます
(基準文との関係は[詳細])
3 私は物理の扱えます
(基準文との関係は[詳細])
4 私は数学の扱えます
(基準文との関係は[詳細])
5 私は相談相手になります
(基準文との関係は[詳細])
6 皆さんの役に立ちたいからです
(基準文との関係は[理由])
(処理8)………………………………
(1) 関連情報の抽出
(質問・問題)に答えてください:検出
(2) 人工知能(出力)
了解しました
(処理9)………………………………
(1) 関連情報の抽出
★ (どの**の○○が**)問題を検出
問題解決のために抽出した関連記録ユニットの情報
番号 1 2 3 4 5
記録ユニット番号 10 11 12 13 14
数値 4000 6000 8000 10000 5000
★ 条件を満足する記録ユニット番号を出力
記録ユニット番号= 10 11 12 14
★ 回答 店Aの予算店Bの予算店Cの予算店Eの予算が8000円以下です
(2) 人工知能(出力)
店Aの予算店Bの予算店Cの予算店Eの予算が8000円以下です
(処理10)………………………………
(1) 関連情報の抽出
★ 記録した情報から選択する問題を検出
★ (どの**の○○が**)問題を検出
問題解決のために抽出した関連記録ユニットの情報
番号 1 2 3 4
記録ユニット番号 5 6 7 9
数値 0.5 1.2 0.8 0.3
★ 条件を満足する記録ユニット番号を出力
記録ユニット番号= 5 7 9
★ 回答 店Aの距離店Cの距離店Eの距離が1km以下です
(2) 人工知能(出力)
店Aの距離店Cの距離店Eの距離が1km以下です
(処理11)………………………………
(1) 関連情報の抽出
★ (どの**の○○が**)問題の検出
問題解決のために抽出した関連記録ユニットの情報
番号 1 2 3
関連ユニット番号 5 7 9
★ 条件を満足する単語番号を出力
回答の単語番号 単語 55 店Cのおすすめ
57 店Eのおすすめ
回答: 店Cのおすすめ店Eのおすすめが日本酒です
(2) 人工知能(出力)
店Cのおすすめ店Eのおすすめが日本酒です
(処理12)………………………………
(1) 関連情報の抽出
問題の回答に対応する検索データ無し
処理状況の表示
処理状況:値段を求める問題検出
処理状況:値段を求めるプログラム起動
処理状況:値段に関する関連情報 検出
処理状況:単価= 100
処理状況:総額= 500
(2) 人工知能(出力)
リンゴ5個の値段は500円です
(処理13)………………………………
(1) 関連情報の抽出
2次微分項の係数= 2
1次微分項の係数= 3
0次項の係数= 1
処理状況:微分方程式の問題解法プログラム起動
Ad2x/_dtdt+Bd_x/_dt+C=0 の解を求める
d_x/_dt(0)=x1,x(0)=x0
A=2 B=3 C=1
x1=5 x0=4
(2) 人工知能(出力)
回答:方程式の解は以下です
x=A exp(p1t)+B exp(p2t)
p1=−1 p2=−0.5
A=−14 B=18
(処理14)………………………………
(1) 関連情報の抽出
問題の回答に対応する検索データ無し
処理状況:交点を求める問題検出
処理状況:関数1の検索起動
処理状況:関数1= f
処理状況:関数1の係数表示 2次=1 1次=2 0次=−4
処理状況:関数2の検索起動
処理状況:関数2= g
処理状況:関数2の係数表示 2次=0 1次=2 0次=0
処理状況:関数(2−1)の係数表示
2次=−1 1次=0 0次=4
処理状況:交点検出
処理状況:X1=−2 X2=2
回答:交点は X1=−2 X2=2 です
(2) 人工知能(出力)
fとgの交点は−2と2です
(処理15)………………………………
(1) 関連情報の抽出
問題の回答に対応する検索データ無し
処理状況:交点を求める問題検出
処理状況:関数1の検索起動
処理状況:関数1= f
処理状況:関数1の係数表示 2次=1 1次=2 0次=−4
処理状況:関数2の検索起動
処理状況:関数2= g
処理状況:関数2の係数表示 2次=0 1次=2 0次=0
処理状況:関数(2−1)の係数表示
2次=−1 1次=0 0次=4
処理状況:交点検出
処理状況:X1=−2 X2=2
処理状況:積分[関数(2−1)]関数の係数表示
3次=−0.333333 2次=0 1次=4
処理状況:★★ 積分範囲[x1−x2]における積分値=10.6667
回答:面積は 10.6667 です
(2) 人工知能(出力)
fとgの囲む面積は10.6667です
(処理16)………………………………
(1) 関連情報の抽出
問題の回答に対応する検索データ無し
処理状況の表示
処理状況:飛距離を求める問題検出
処理状況:飛距離を求めるプログラム起動
処理状況:初速検索起動
処理状況:迎角検索起動
処理状況:初速情報の検出 v =10m/s
処理状況:迎角情報の検出 t =45deg
処理状況:飛距離の導出(数値) 10.2041m
回答:飛距離は 10.2041 mです
(2) 人工知能(出力)
飛距離は 10.2041 mです
(処理17)………………………………
(1) 関連情報の抽出
★ (何が**ですか)問題の検出
問題解決のために抽出した関連記録ユニットの情報
番号 1 2 3
記録ユニット番号 71 72 73
右項型 3 4 5
右項[3] − − +
右項[2] k d m_
右項[1] x x g
回答:バネの力が ダンパーの力が 重力の力が 働きます
(2) 人工知能(出力)
バネの力がダンパーの力が重力の力が働きます
(注)項の型定義
0:データ無し、1:記号、2:±記号、3:±記号・x、
4:±記号・d_x/_dt、5:±記号・記号、6:±d2x/_dtdt
(処理18)………………………………
(1) 関連情報の抽出
★ 型番号と記号番号で表示
左項数= 1
右項数= 3
左項型 6 0 0
左項[1] 37 0 0
左項[2] 142 0 0
左項[3] 91 0 0
左項[4] 40 0 0
左項[5] 28 0 0
右項型 3 4 5
右項[1] 38 38 37
右項[2] 78 71 142
右項[3] 91 91 74
右項[4] 40 40 40
右項[5] 28 28 28
★ 型番号と記号で表示
左項型 6 0 0
左項[1] +
左項[2] m_
左項[3] x
左項[4] /
左項[5] 1
右項型 3 4 5
右項[1] − − +
右項[2] k d m_
右項[3] x x g
右項[4] / / /
右項[5] 1 1 1
第2ステップ (移項処理)……………………
★ 型番号と記号番号で表示
左項数= 2
右項数= 2
左項型 6 4 0
左項[1] 37 37
左項[2] 142 71
左項[3] 91 91
左項[4] 40 40
左項[5] 28 28
右項型 3 0 5
右項[1] 38 0 37
右項[2] 78 0 142
右項[3] 91 0 74
右項[4] 40 0 40
右項[5] 28 0 28
★ 型番号と記号で表示
左項型 6 4
左項[1] + +
左項[2] m_ d
左項[3] x x
左項[4] / /
左項[5] 1 1
右項型 3 0 5
右項[1] − +
右項[2] k m_
右項[3] x g
右項[4] / /
右項[5] 1 1
★ 型番号と記号番号で表示
左項数= 3
右項数= 1
左項型 6 4 3
左項[1] 37 37 37
左項[2] 142 71 78
左項[3] 91 91 91
左項[4] 40 40 40
左項[5] 28 28 28
右項型 0 0 5
右項[1] 0 0 37
右項[2] 0 0 142
右項[3] 0 0 74
右項[4] 0 0 40
右項[5] 0 0 28
★ 型番号と記号で表示
左項型 6 4 3
左項[1] + + +
左項[2] m_ d k
左項[3] x x x
左項[4] / / /
左項[5] 1 1 1
右項型 0 0 5
右項[1] +
右項[2] m_
右項[3] g
右項[4] /
右項[5] 1
第3ステップ (正規化処理)……………………
左項型=6AND定数項>0の場合の定数項番号=142
★ 型番号と記号番号で表示
左項数= 3
右項数= 1
左項型 6 4 3
左項[1] 37 37 37
左項[2] 28 71 78
左項[3] 91 91 91
左項[4] 40 40 40
左項[5] 28 142 142
右項型 0 0 5
右項[1] 0 0 37
右項[2] 0 0 28
右項[3] 0 0 74
右項[4] 0 0 40
右項[5] 0 0 28
★ 型番号と記号で表示
左項型 6 4 3
左項[1] + + +
左項[2] 1 d k
左項[3] x x x
左項[4] / / /
左項[5] 1 m_ m_
右項型 0 0 5
右項[1] +
右項[2] 1
右項[3] g
右項[4] /
右項[5] 1
★ 型番号と記号で表示
左項型 6 4 3
左項[1] + + +
左項[2] m_ d k
左項[3] x x x
左項[4] / / /
左項[5] 1 1 1
右項型 0 0 5
右項[1] +
右項[2] m_
右項[3] g
右項[4] /
右項[5] 1
第3ステップ (正規化処理)……………………
左項型=6AND定数項>0の場合の定数項番号=142
★ 型番号と記号番号で表示
左項数= 3
右項数= 1
左項型 6 4 3
左項[1] 37 37 37
左項[2] 28 71 78
左項[3] 91 91 91
左項[4] 40 40 40
左項[5] 28 142 142
右項型 0 0 5
右項[1] 0 0 37
右項[2] 0 0 28
右項[3] 0 0 74
右項[4] 0 0 40
右項[5] 0 0 28
★ 型番号と記号で表示
左項型 6 4 3
左項[1] + + +
左項[2] 1 d k
左項[3] x x x
左項[4] / / /
左項[5] 1 m_ m_
右項型 0 0 5
右項[1] +
右項[2] 1
右項[3] g
右項[4] /
右項[5] 1
第4ステップ (運動方程式の表示)………………
回答:運動方程式は以下です
g =9.8
m_=1
k =1
d =1
d2x/_dtdt+(d/m_)d_x/_dt+(k/m_)x = g
(2) 人工知能(出力)
運動方程式は以下です
g =9.8
m_=1
k =1
d =1
d2x/_dtdt+(d/m_)d_x/_dt+(k/m_)x = g
An example of the operation of this artificial intelligence is shown below. Explain that you can solve conversations and problems using information given by humans. Information (source) necessary to answer this question was taught to artificial intelligence in advance. In the next conversation, artificial intelligence uses the knowledge to respond to human conversations and questions. Artificial intelligence detects characteristic patterns in each conversation, autonomously activates processing programs necessary for problem recognition, identification, and problem solving, and newly generated information is recorded as knowledge in a pattern recorder. Is expanding. The details of the artificial intelligence process in each conversation are shown in (Process 1) to (Process 18).
NO (conversation number) Conversation between human and artificial intelligence 1 Human: Nice to meet you 2 Artificial intelligence: I am an atom (Process 1)
3 I am artificial intelligence 4 Human: I am Taro 5 Artificial intelligence: My profession is assistant (Process 2)
6 What is Taro's occupation 7 Human: My occupation is a teacher 8 Artificial intelligence: Teachers are difficult (Process 3)
9 Human: I teach physics in class 10 Human: I teach tennis in after-school activities 11 Artificial Intelligence: What is Taro's hobby (Process 4)
12 Human: My hobby is sports 13 Artificial intelligence: Speaking of sports (Process 5)
14 Kay won the tennis tournament yesterday 15 Kay won straight in the final 16 It ’s amazing 17 Human: Why is the profession assistant 18 Artificial intelligence: Because I want to help you
19 Human: Please tell me more about the assistant. 20 Artificial Intelligence: I support problem solving (Process 7)
21 I can handle life problems 22 I can handle physics problems 23 I can handle math problems 24 I can be a consultant 25 Human: Then answer the questions 26 Artificial intelligence: OK (Process 8)
27 Human: Life problems 28 Human: Tell me about the restaurant 29 Human: Which store budget is 8000 yen or less 30 Artificial intelligence: Budget of store A Budget store B of budget store C Budget of store E 8000 yen or less (Process 9)
31 Human: Further, which store distance is 1 km or less? 32 Artificial intelligence: Distance of store A Distance of store C Distance of store E is less than 1 km
33 Human: Which store is recommended for sake? 34 Artificial intelligence: Recommended for store C, recommended store E is sake (Process 11)
35 Human: Life problems 36 Human: The price of 3 apples is 300 yen 37 Human: What is the price of 5 apples 38 Artificial intelligence: The price of 5 apples is 500 yen (Process 12)
39 Human: Mathematics problem 40 Human: x0 is 4 41 Human: x1 is 5 42 Human: [2 d2x / _dtdt + 3 d_x / _dt + 1 = 0]
43 Artificial intelligence: The solution of the equation is as follows (Process 13)
x = A exp (p1t) + B exp (p2t)
p1 = -1 p2 = -0.5
A = -14 B = 18
44 Human: Mathematics problem 45 Human: [f = x ** 2 + 2x-4]
46 Human: [g = 2x]
47 Human: What is the intersection of f and g 48 Artificial intelligence: The intersection of f and g is -2 and 2 (Process 14)
49 Human: What is the area enclosed by f and g 50 Artificial intelligence: The area enclosed by f and g is 10.66667 (Process 15)
51 Human: It is a physical problem 52 Human: The initial velocity of the ball is v 53 Human: The angle of attack of the ball is t 54 Human: v is 10 m / s 55 Human: t is 45 deg 56 Human: The flying distance of the ball 57 Artificial intelligence: Flight distance is 10.2041 m (Process 16)
58 Human: Physical problem 59 Human: Spring force is -kx 60 Human: Damper force is -d dx / dt 61 Human: Gravity force is + m_g 62 Human: Spring force works on object C 63 Human: The force of the damper works on the object C 64 Human: The force of gravity works on the object C 65 Human: What works on the object C 66 Artificial intelligence: The force of the spring uses the force of the damper of gravity Power works (Process 17)
67 Human: g is 9.8 68 Human: m_ is 1 69 Human: k is 1 70 Human: d is 1 71 Human: What is the equation of motion? 72 Artificial intelligence: The equation of motion is ( Process 18)
g = 9.8
m_ = 1
k = 1
d = 1
d2x / _dtdt + (d / m_) d_x / _dt + (k / m_) x = g
★★★★★ The following is the processing of artificial intelligence (details) ★★★★★
(Process 1) ………………………………
(1) Extraction of related information Greetings (Nice to meet) Information (name, occupation) of detected speaker (partner) Collection program start Speaker (partner) information (name) not recorded (2) Artificial intelligence (output)
I am an atom I am an artificial intelligence (Process 2) ………………………………
(1) Extracting related information Detecting speaker (partner) information (name) Talker (partner) information (name) is Taro speaker (partner) information (name) is already recorded speaker (partner) Information (occupation) is not recorded (2) Artificial intelligence (output)
My occupation is assistant What is Taro's occupation (Process 3) ………………………………
(1) Extraction of related information Speaker's (partner) information (name) is already recorded speaker's (partner) information (name) is Taro's profession answer Speaker's (partner) information (profession) is teacher Search and output the opinions of the speaker (partner) about their occupation (2) Artificial intelligence (output)
Teacher is hard (Process 4) ………………………………
(1) Extraction of related information Speaker (partner) information (hobby) Collecting program start Speaker (partner) information (hobby) not recorded (2) Artificial intelligence (output)
What is Taro's hobby? (Process 5) ………………………………
(1) Extracting related information Detecting speaker (partner) information (hobby) Speaker (partner) information (hobby) Searching for and outputting related news related to sports speaker (partner) hobby Analysis (Evaluation of related information that detects and detects “Straight Win” (Evaluation of “Straight Win” and evaluation of related information that is detected and output) (2) Artificial Intelligence (Output)
Speaking of sports, Kay won the tennis tournament yesterday. Kay is amazing that he won straight in the final round (Process 6) ………………………………
(1) Extraction of related information qqtype = 6; // Set the type of question to (why is **) There is a connection relationship: Number of connections = 6
senconrel [i] [j] = 1
senconrel [i] [j] = 1
senconrel [i] [j] = 1
senconrel [i] [j] = 1
senconrel [i] [j] = 1
senconrel [i] [j] = 5
There is a reason item in the connection relationship: statement number of the connection destination in the detection connection relationship (reason) = 28
Text output of connection (reason): Because I want to help you (2) Artificial intelligence (output)
I want to help you (Process 7) ………………………………
★★ Accessing related information using connection between sentences (example) ★★
Displaying the standard sentence
My profession is an assistant. The number of standard sentences and related sentences (connections) is 6.
1 I will help solve the problem
(Relationship with standard text is [Details])
2 I can handle life
(Relationship with standard text is [Details])
3 I can handle physics
(Relationship with standard text is [Details])
4 I can handle mathematics
(Relationship with standard text is [Details])
5 I will consult
(Relationship with standard text is [Details])
6 Because I want to help you
(Relationship with standard statement is [reason])
(Process 8) ………………………………
(1) Answer related information extraction (questions / problems): detection (2) artificial intelligence (output)
I understand (Process 9) ………………………………
(1) Extraction of related information ★ ( ** of ** is ** ) Detecting a problem Information number of a related recording unit extracted for solving a problem 1 2 3 4 5
Recording unit number 10 11 12 13 14
Numerical value 4000 6000 8000 10000 5000
★ The recording unit number that satisfies the conditions is the output recording unit number = 10 11 12 14
★ Answer The budget of budget store E of budget store B of store A is less than 8000 yen (2) Artificial intelligence (output)
Budget of store A budget store B of budget store C of budget store C is 8000 yen or less (Process 10) ………………………………
(1) Extraction of related information ★ Detects a problem to be selected from recorded information ★ (which ** is ** ) Detects a problem Information number of a related recording unit extracted for solving a problem 1 2 3 4
Recording unit number 5 6 7 9
Numerical value 0.5 1.2 0.8 0.3
★ The recording unit number that satisfies the conditions is the output recording unit number = 5 7 9
★ Answer The distance of distance store E of store A is 1 km or less (2) Artificial intelligence (output)
The distance of the distance store E of the distance store C of the store A is 1 km or less (processing 11) ………………………………
(1) Advice extraction ★ (which ** in ○○ is **) of the extracted associated recording unit for detection problem solving problems information number 1 2 3
Related unit number 5 7 9
★ Outputs the word number that satisfies the conditions Word number of the answer Word 55 Recommendation of shop C
57 Recommended Responses from Store E: Recommended at Store C Recommended Store E is Sake (2) Artificial Intelligence (Output)
The recommendation of store E at store C is sake (Process 12) ………………………………
(1) Processing status display processing status of no search data corresponding to answer of related information extraction problem: Problem detection processing status for obtaining price: Program start processing status for obtaining price: Related information regarding price Detection processing status: Unit price = 100
Processing status: Total = 500
(2) Artificial intelligence (output)
The price of 5 apples is 500 yen (process 13).
(1) Extraction of related information Coefficient of second derivative term = 2
First derivative term coefficient = 3
0th order coefficient = 1
Processing status: Differential equation problem solving program launched
Find the solution Ad2x / _dtdt + Bd_x / _dt + C = 0
d_x / _dt (0) = x1, x (0) = x0
A = 2 B = 3 C = 1
x1 = 5 x0 = 4
(2) Artificial intelligence (output)
Answer: The solution of the equation is x = A exp (p1t) + B exp (p2t)
p1 = -1 p2 = -0.5
A = -14 B = 18
(Process 14) ………………………………
(1) Processing status without search data corresponding to the answer to the related information extraction problem: Problem detection processing status for finding an intersection: Search start processing status of function 1: Function 1 = f
Processing status: coefficient display of function 1 2nd order = 1 1st order = 2 0th order = -4
Processing status: Function 2 search activation processing status: Function 2 = g
Processing status: Function 2 coefficient display 2nd order = 0 1st order = 2 0th order = 0
Processing status: Coefficient display of function (2-1)
2nd order = -1 1st order = 0 0th order = 4
Processing status: intersection detection processing status: X1 = −2 X2 = 2
Answer: The intersection is X1 = -2 X2 = 2 (2) Artificial intelligence (output)
The intersection of f and g is -2 and 2 (Process 15) ………………………………
(1) Processing status without search data corresponding to the answer to the related information extraction problem: Problem detection processing status for finding an intersection: Search start processing status of function 1: Function 1 = f
Processing status: coefficient display of function 1 2nd order = 1 1st order = 2 0th order = -4
Processing status: Function 2 search activation processing status: Function 2 = g
Processing status: Function 2 coefficient display 2nd order = 0 1st order = 2 0th order = 0
Processing status: Coefficient display of function (2-1)
2nd order = -1 1st order = 0 0th order = 4
Processing status: intersection detection processing status: X1 = −2 X2 = 2
Processing status: integral [function (2-1)] function coefficient display 3rd order = −0.333333 2nd order = 0 1st order = 4
Processing status: ★★ Integration value in integration range [x1-x2] = 10.6667
Answer: The area is 10.66667 (2) Artificial intelligence (output)
The area enclosed by f and g is 10.66667 (Process 16) ………………………………
(1) Display processing status of no search data corresponding to answer of related information extraction problem Processing status: Problem detection processing status for determining flight distance: Program startup processing status for determining flight distance: Initial speed search startup processing status: Angle of attack search Startup processing status: detection of initial speed information v = 10 m / s
Processing status: detection of angle-of-attack information t = 45 deg
Processing status: Derivation of flight distance (numerical value) 10.2041m
Answer: The flight distance is 10.2041 m. (2) Artificial intelligence (output)
The flight distance is 10.2041 m (Process 17) ………………………………
(1) Extraction of related information ★ (What is ** ) Detection of problem Information number of related recording unit extracted to solve problem 1 2 3
Recording unit number 71 72 73
Right-term type 3 4 5
Right term [3]--+
Right term [2] k d m_
Right term [1] x x g
Answer: Spring force Damper force Gravity force (2) Artificial intelligence (Output)
The spring force is the damper force, and the gravity force is the gravity force. Note type definition 0: No data, 1: Symbol, 2: ± symbol, 3: ± symbol, x,
4: ± symbol / d_x / _dt, 5: ± symbol / symbol, 6: ± d2x / _dtdt
(Process 18) ………………………………
(1) Extraction of related information ★ Displayed with model number and symbol number Left number = 1
Number of right terms = 3
Left term type 6 0 0
Left term [1] 37 0 0
Left term [2] 142 0 0
Left term [3] 91 0 0
Left term [4] 40 0 0
Left term [5] 28 0 0
Right-term type 3 4 5
Right [1] 38 38 37
Right [2] 78 71 142
Right [3] 91 91 74
Right term [4] 40 40 40
Right [5] 28 28 28
★ Displayed with model number and symbol Left term type 6 0 0
Left term [1] +
Left term [2] m_
Left term [3] x
Left item [4] /
Left term [5] 1
Right-term type 3 4 5
Right term [1]--+
Right term [2] k d m_
Right term [3] x x g
Right [4] / / /
Right [5] 1 1 1
Second step (shift process) ……………………
★ Displayed by model number and symbol number Number of items left = 2
Number of right terms = 2
Left term type 6 4 0
Left term [1] 37 37
Left term [2] 142 71
Left item [3] 91 91
Left term [4] 40 40
Left term [5] 28 28
Right-term type 3 0 5
Right term [1] 38 0 37
Right term [2] 78 0 142
Right [3] 91 0 74
Right term [4] 40 0 40
Right [5] 28 0 28
★ Display with model number and symbol Left term type 6 4
Left term [1] + +
Left term [2] m_d
Left term [3] x x
Left item [4] / /
Left term [5] 1 1
Right-term type 3 0 5
Right term [1]-+
Right term [2] km_
Right term [3] x g
Right [4] / /
Right [5] 1 1
★ Displayed by model number and symbol number Number of left items = 3
Number of right terms = 1
Left term type 6 4 3
Left term [1] 37 37 37
Left term [2] 142 71 78
Left item [3] 91 91 91
Left term [4] 40 40 40
Left term [5] 28 28 28
Right term type 0 0 5
Right term [1] 0 0 37
Right term [2] 0 0 142
Right term [3] 0 0 74
Right term [4] 0 0 40
Right [5] 0 0 28
★ Displayed by model number and symbol Left-hand type 6 4 3
Left term [1] + + +
Left term [2] m_d k
Left term [3] x x x
Left item [4] / / /
Left term [5] 1 1 1
Right term type 0 0 5
Right [1] +
Right [2] m_
Right term [3] g
Right [4] /
Right [5] 1
3rd step (Normalization) ……………………
Left term type = 6 AND constant term> 0 when constant term number = 142
★ Displayed by model number and symbol number Number of left items = 3
Number of right terms = 1
Left term type 6 4 3
Left term [1] 37 37 37
Left term [2] 28 71 78
Left item [3] 91 91 91
Left term [4] 40 40 40
Left item [5] 28 142 142
Right term type 0 0 5
Right term [1] 0 0 37
Right term [2] 0 0 28
Right term [3] 0 0 74
Right term [4] 0 0 40
Right [5] 0 0 28
★ Displayed by model number and symbol Left-hand type 6 4 3
Left term [1] + + +
Left term [2] 1 d k
Left term [3] x x x
Left item [4] / / /
Left term [5] 1 m_ m_
Right term type 0 0 5
Right [1] +
Right term [2] 1
Right term [3] g
Right [4] /
Right [5] 1
★ Displayed by model number and symbol Left-hand type 6 4 3
Left term [1] + + +
Left term [2] m_d k
Left term [3] x x x
Left item [4] / / /
Left term [5] 1 1 1
Right term type 0 0 5
Right [1] +
Right [2] m_
Right term [3] g
Right [4] /
Right [5] 1
3rd step (Normalization) ……………………
Left term type = 6 AND constant term> 0 when constant term number = 142
★ Displayed by model number and symbol number Number of left items = 3
Number of right terms = 1
Left term type 6 4 3
Left term [1] 37 37 37
Left term [2] 28 71 78
Left item [3] 91 91 91
Left term [4] 40 40 40
Left item [5] 28 142 142
Right term type 0 0 5
Right term [1] 0 0 37
Right term [2] 0 0 28
Right term [3] 0 0 74
Right term [4] 0 0 40
Right [5] 0 0 28
★ Displayed by model number and symbol Left-hand type 6 4 3
Left term [1] + + +
Left term [2] 1 d k
Left term [3] x x x
Left item [4] / / /
Left term [5] 1 m_ m_
Right term type 0 0 5
Right [1] +
Right term [2] 1
Right term [3] g
Right [4] /
Right [5] 1
4th step (Display of equation of motion) ………………
Answer: The equation of motion is g = 9.8
m_ = 1
k = 1
d = 1
d2x / _dtdt + (d / m_) d_x / _dt + (k / m_) x = g
(2) Artificial intelligence (output)
The equation of motion is g = 9.8
m_ = 1
k = 1
d = 1
d2x / _dtdt + (d / m_) d_x / _dt + (k / m_) x = g

以下では本人工知能装置の特徴について説明する。
本人工知能の最大の特徴は解の導出過程に関して透明性が非常に高いということである。本人工知能の各記録ユニットの情報(照合および合致(起動)の条件、パターン、パターン間の接続関係、接続関係のあるパターンとの関係性およびパターンの分析結果)は全て人間が理解できる言語、記号等をパターン化して表現したものである。また各記録ユニットと接続関係にある記録ユニットおよび接続の関係性についても明確であるので論理展開についても容易にトレースすることができる。本人工知能は全ての処理過程がモニター可能であり、かつ人間が理解可能な形式、構造を有した透明性の高いシステムである。つまり本人工知能は処理過程の妥当性および安全性を人間が理解し、確認できるシステムとなっている。また、本人工知能内に構築する知識体系は真実、事実に基づき人間の道徳、規範に反していないことを逐次、確認しながら構築していったものである。人間の道徳、規範に反したものは逐次、検出し排除するので誤った知識(人間の道徳、規範に反した知識)は組み込まれない。人工知能の自律化、機能、性能を高めていく段階で人間が設定した規則、常識、道徳を逸脱しないことを管理することができる。
また人工知能の中で構築される知識体系が現実に整合して処理できるかという「記号接地問題」に関しては、本人工知能の知識体系は人間が教示する現実に対応する「言語、文、文章、記号、数式、物理式、化学式」を源泉データとし、それを逐次、パターンに変換し、情報間の関係をパターン間の接続関係として構造化したものであるので、本質的に入力した情報に対応する「現実」と等価であり、整合性が取れたものとなっている。
本人工知能は人間が築いてきた知性、知識を言語、文、文章等の情報として逐次、入力・評価していき、安全で価値があると判断した情報を既に記録している知識との関係を分析して知識体系に自律的に組み込んでいく。入力情報を既に知識体系に記録している情報と比較・評価し、安全で価値があると判断した情報に分析結果を付加し、さらに発展させるという方式で、知識体系を自己組織的に構築していく。このように構築した知識体系は人間の知性に相当するデータベースとなっており、問題の解決および人間との会話に使用することができる。
本人工知能は構築した知識体系を活用し、与えられた問題に関する関連データを検索する。検索された関連データと関係を有する情報に関しても、接続先、関係性を識別することができる。処理の過程(考え方のプロセス)において適切な関係性のものを逐次、選択していくことにより、複数の関連情報から適切なものに絞り込み、思考プロセスを適切な目標に向けて自律的に進めていくことが可能である。
本人工知能は真実、事実、規則、常識、定義、論理に相当する情報を逐次入力していくことで、情報と情報の接続関係を自律的に生成していく。情報を入力すると情報に対応する数々のパターンが励起し、パターンの集合(概念)とパターンの集合(概念)との間の接続関係が人間の指示または学習により強化していく。文、文章は単独で存在するのではなく、互いに関係を有しており文脈(コンテキスト)の中で位置づけされる。後に出現する文は前に出現した文の影響を受けている。この事を反映するメカニズムとして本人工知能は入力した文に応じて対応するパターンが励起し、励起したパターンの接続関係記録部には、当パターンが励起する前に励起したパターンの履歴が記録され、関連するパターンとの接続関係が逐次強化する機能を組み込んでいる。この機能により、文に対応するパターンは全て関連する文との関係、つまり文脈を踏まえて励起するように学習が進んでいく動作を実現することができる。多数の文、文章を入力することにより文脈のネットワークが自律的に生成され、ある概念(文、文章)が入力されると、その概念は生成された文脈に沿って遷移していくことになる。
このような情報間の接続関係を真実、事実、規則、常識、定義、論理に相当する情報について実施していくと、真実、事実、規則、常識、定義、論理に相当する知識体系を自律的に生成することができる。学習が進むと本人工知能は入力された情報が既に記録している情報と整合しているが、整合していないかの識別が可能となる。つまり、真実、事実、規則、常識、定義、論理に相当する知識、知性を獲得したことになる。情報(思考)から情報(思考)への遷移は生成した文脈に沿って実施することも可能であるし、情報に含まれたキーワードを使用して連想記憶的に情報を検索することにより思考を遷移させることも可能である。思考パターンの多様な遷移方法を組み合わせることにより人間に近い会話を実施することもできる。
また質問に対する回答も知識体系から関連情報を検索して生成するのみならず、質問の意味を解釈した上で関連情報を検索し、検索したデータを処理することにより質問に対する回答を生成して答えるという動作ができる。これは質問のタイプを識別し、質問の回答に対応する情報を検索し、質問の回答に対応する情報が無い場合には、質問の回答に対応する情報を生成するプログラムを自律的に起動し処理を行うことにより実現できる。質問の回答に対応する情報を生成するための情報を知識体系から検索し、その情報が無ければ、さらにその情報を生成するためのプログラムを起動して必要な情報を生成していくという逐次処理を行うことにより実現する。
人間の思考パターンの遷移は条件により遷移先が変わるという観点から、一般的に条件付き処理として表現することができる。本人工知能は言語の意味を解釈し、必要に応じて自律的に条件付処理に変換する。条件つき処理の条件の詳細は、該当する言語から検索用のパターンを生成し、自律的に検索する。検索した情報が条件を満足するか否かの判断を実施し、満足する場合は該当する処理を実行する。人間が知識として使用している問題解決策を、そのまま言語として入力しても、その意味を逐次、解釈し条件付処理を自律的に進めていき問題を解決していく。処理の途中で新規の情報が必要な場合は、その情報要求を通知し、該当の情報が獲得されると、その内容に応じた処理を実施する。
人間の指示により情報(パターン)間の接続関係を生成する場合はある概念(文、文章)に対応するパターンを励起し、つぎの概念(文、文章)のパターンを励起することにより実施することができる。先に起動したパターンから後で起動したパターンに向け、遷移するように接続関係を生成するように設定しておけば、概念と概念の間の接続関係を逐次、生成していき、次に同様の概念が励起すると、生成した接続関係に沿ってパターンが自動的に励起するようになる。遷移先が複数の場合は、状況に適した関係性を人間の指示または自律的に設定し、複数の遷移先の候補から状況に適した関係性のものが選択されるようにする。なお、関係性の自律的設定は思考パターンの遷移の目的、状況を分析し適切な関係性を設定する機能により実施できる。
Below, the feature of this artificial intelligence device is explained.
The biggest feature of this artificial intelligence is that the solution derivation process is very transparent. Information of each recording unit of this artificial intelligence (matching and matching (activation) conditions, patterns, connection relations between patterns, relations with patterns with connection relations and pattern analysis results) are all human language understandable, Symbols and the like are expressed as patterns. Further, since the relationship between the recording unit and the connection relationship with each recording unit is also clear, the logical development can be easily traced. This artificial intelligence is a highly transparent system that can monitor all processes and has a format and structure that can be understood by humans. In other words, this artificial intelligence is a system that allows humans to understand and confirm the validity and safety of the process. In addition, the knowledge system built in this artificial intelligence was constructed by confirming that it is not contrary to human morals and norms based on truth and fact. Since things that violate human morals and norms are successively detected and eliminated, false knowledge (knowing that violates human morals and norms) is not incorporated. It is possible to manage not to deviate from the rules, common sense and morals set by humans at the stage of increasing the autonomy, function and performance of artificial intelligence.
In addition, regarding the “symbol grounding problem” of whether the knowledge system constructed in artificial intelligence can be processed consistently with the reality, the knowledge system of this artificial intelligence corresponds to the “language, sentence, sentence” corresponding to the reality taught by humans. `` Symbols, mathematical formulas, physical formulas, chemical formulas '' are used as source data, which are sequentially converted into patterns, and the relationships between information are structured as connection relationships between patterns. It is equivalent to the corresponding “reality” and is consistent.
This artificial intelligence inputs and evaluates human intelligence and knowledge as language, sentences, sentences, etc., and relations with knowledge that has already been recorded as safe and valuable information Are analyzed and incorporated into the knowledge system autonomously. Comparing and evaluating input information with information already recorded in the knowledge system, adding analysis results to information judged safe and valuable, and further developing the knowledge system in a self-organizing manner To go. The knowledge system constructed in this way is a database corresponding to human intelligence, and can be used for problem solving and conversation with humans.
This artificial intelligence utilizes the constructed knowledge system and searches for related data on a given problem. The connection destination and the relationship can also be identified for the information related to the retrieved related data. By selecting the relevant ones in the processing process (thinking process) one after another, the relevant information is narrowed down to the appropriate one, and the thinking process proceeds autonomously toward the appropriate goal. It is possible to go.
This artificial intelligence autonomously generates information and information connection relationships by sequentially inputting information corresponding to truth, facts, rules, common sense, definition, and logic. When information is input, a number of patterns corresponding to the information are excited, and the connection relationship between the pattern set (concept) and the pattern set (concept) is strengthened by human instruction or learning. Sentences and sentences do not exist alone, but are related to each other and positioned in the context. The sentence that appears later is affected by the sentence that appears earlier. As a mechanism to reflect this, in this artificial intelligence, the corresponding pattern is excited according to the input sentence, and the history of the excited pattern is recorded in the connection relation recording part of the excited pattern before the pattern is excited. Incorporates a function that gradually strengthens the connection relationship with related patterns. With this function, it is possible to realize an operation in which learning proceeds so that all patterns corresponding to a sentence are excited based on the relationship with a related sentence, that is, context. By entering a large number of sentences and sentences, a context network is generated autonomously. When a certain concept (sentence or sentence) is input, the concept transitions along the generated context. .
When such information is connected to information corresponding to truth, facts, rules, common sense, definition, and logic, the knowledge system corresponding to truth, facts, rules, common sense, definition, and logic is autonomous. Can be generated. As learning proceeds, the artificial intelligence can identify whether the input information is consistent with the information already recorded, but not. In other words, it has acquired knowledge and intelligence equivalent to truth, facts, rules, common sense, definition and logic. The transition from information (thinking) to information (thinking) can be carried out according to the generated context, and thinking can be done by searching for information in an associative memory using keywords included in the information. It is also possible to make a transition. By combining various transition methods of thought patterns, it is possible to carry out a conversation close to a human being.
Answers to questions are generated not only by searching related information from the knowledge system, but also by searching for related information after interpreting the meaning of the question and processing the searched data to generate and answer the question. Can be operated. This identifies the type of question, searches for information corresponding to the answer to the question, and if there is no information corresponding to the answer to the question, it autonomously launches a program that generates information corresponding to the question answer. This can be realized by processing. Sequential processing of retrieving information for generating information corresponding to the answer of the question from the knowledge system, and if there is no such information, launching a program for generating that information and generating the necessary information It is realized by doing.
The transition of human thought patterns can be generally expressed as a conditional process from the viewpoint that the transition destination changes depending on the condition. This artificial intelligence interprets the meaning of the language and autonomously converts it into conditional processing as needed. As for the details of the conditional processing conditions, a search pattern is generated from the corresponding language and searched autonomously. Judgment is made as to whether or not the retrieved information satisfies the condition. If the information is satisfied, the corresponding processing is executed. Even if a problem solution that humans use as knowledge is input as a language as it is, the meaning is sequentially interpreted and conditional processing proceeds autonomously to solve the problem. When new information is required during the process, the information request is notified, and when the corresponding information is acquired, the process according to the content is performed.
When a connection relationship between information (patterns) is generated by a human instruction, the pattern corresponding to a concept (sentence, sentence) is excited and the pattern of the next concept (sentence, sentence) is excited. Can do. If it is set so that the connection relationship is generated so as to transition from the pattern activated first to the pattern activated later, the connection relationship between concepts will be generated sequentially, and then the same When this concept is excited, the pattern is automatically excited along the generated connection relation. When there are a plurality of transition destinations, a relationship suitable for the situation is set by a human instruction or autonomously, and a relationship suitable for the situation is selected from a plurality of transition destination candidates. In addition, the autonomous setting of the relationship can be performed by a function of analyzing the purpose and situation of the transition of the thinking pattern and setting an appropriate relationship.

本人工知能は汎用であるため数々の応用分野が考えられる。ここでは本人工知能の特徴に焦点を当てた一部について説明する。ここではAIと表現しているがAIを搭載したロボット、機器への適用も可能である。
◇パーソナルAI
人工知能へ入力する情報を個人的にカスタマイズし、個人専用のAIを構築する。(個人の思考・思想の記録、話し相手、相談相手、アシスタント、パートナー)
◇汎用AI
規則、常識、道徳を学習した人工知能に専門知識(数学、物理、化学、法律、設計、一般科学等)を学習させ、様々な社会活動において人間を支援する。
◇大規模AI
上記の汎用AIの専門能力を高め、複数の専門分野AIとマネージメントAIからなるネットワークを構成し協調させ、難度の高い問題、複数の専門知識が必要な問題解決を支援する。
◇趣味、ゲームAI
関心のあるテーマ、分野の情報を人工知能に収集、整理、報告させる。人工知能の知識、能力を向上させる育成ゲーム。人工知能が獲得した知識、能力をサークルメンバーで共有、交換等。
Since this artificial intelligence is general-purpose, there are many possible applications. Here, a part focusing on the features of this artificial intelligence will be explained. Although it is expressed as AI here, it can also be applied to robots and devices equipped with AI.
◇ Personal AI
Personalize the information input to artificial intelligence and build a personal AI. (Record of personal thoughts and thoughts, talking partner, consultant, assistant, partner)
◇ General purpose AI
Learn artificial knowledge (math, physics, chemistry, law, design, general science, etc.) from artificial intelligence that has learned rules, common sense, and morals, and support humans in various social activities.
◇ Large-scale AI
The specialized ability of the above-mentioned general-purpose AI is enhanced, and a network composed of a plurality of specialized fields AI and a management AI is formed and coordinated to support problem solving that requires a high degree of difficulty and a plurality of specialized knowledge.
◇ Hobby, Game AI
Let artificial intelligence collect, organize, and report information on topics and fields of interest. A game to improve the knowledge and ability of artificial intelligence. Sharing and exchanging knowledge and abilities acquired by artificial intelligence among circle members.

発明の効果Effect of the invention

本発明によれば、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのかを分析し文構造に整理するとともに有用と判断した情報を自律的に記録し知識体系として構築することが可能である。また入力した情報を内部に構築した知識体系と照合して評価し、評価結果に応じた処理(情報の記録、知識体系の更新・改良、指示内容の実行、質問に対する回答)を自律的に実施する。さらに情報に含まれている単語の一部を一般化し、逐次入力される情報と情報の関係を情報に対応するするパターンの集合とパターンの集合との関係を強化することにより有意なパターン間の関係を抽出し、一連の入力情報から常識、一般的な考え方および問題解決方法を自律的に構築していくことが可能である。
従来は機械に入力情報の処理等を行わせる場合、計算機にあらかじめプログラムを設定する必要があった。入力した情報から状況を判断するプログラム、個々の条件に応じ機械に動作させるプログラムを作成し、機械に搭載した計算機に実装し実行する必要があった。プログラムは専用のプログラム言語により作成する必要があり、開発に多大な時間を要する等のデメリットがあった。条件の検出および対応する動作が適切でなければ、計算機に実装したプログラムを人間が修正する必要があり、修正に多大な時間を要する等のデメリットがあった。
本発明によれば情報の処理の仕方を学習することにより処理を実施するので、逐次プログラミングする必要は無い。処理の方法に関する情報を自然言語で入力することにより、機械が処理の方法を記録、学習していくので大幅に労力を削減することができる。
また、処理の変更もプログラムの変更によらず、パターンの変更とパターン間の接続を変更することにより対応することが可能であるので非常に柔軟性、対応力の高いシステムとなっている。
通常のプログラミングでは変数、定数の定義および処理の内容を事前に決めておく必要があるため、処理結果に応じて多くの分岐が生じるような処理の実現には多大な労力を要する。(現実的には実現が困難な場合が多い。)一方、本発明では処理に使用する変数および定数に対応する情報の検索は自然言語(単語、記号)を組み合わせたパターンを使用することにより実施しており、より人間の思考に近い情報の検索方式となっている。所謂、連想記憶システムにおける情報の記録および検索を具体化したシステムとなっているため、処理に使用する変数、定数に対応する情報の格納場所を事前に厳密に定義しておく必要は無い。新規の情報が発生した段階で自然言語(単語、記号)の組み合わせたパターンを検索キーとして逐次、登録・記録することができる。また登録・記録した情報を自然言語(単語、記号)の組み合わせた検索パターンで容易に検索することができる。これによりプログラミングにおける変数および定数の定義という制約に縛られないシステムとなっている。
さらに、処理の内容については条件付き処理を状況(パターンおよびパターンの分析結果により条件の成立性を識別する)に応じて条件に対応する処理を実施し、実施した処理において、さらに条件付き処理が出現すれば、その条件を分析し処理を決定するという方式で逐次処理を進めていくことが可能である。従って、人間が問題解決するのと同じ思考プロセス(条件文の集合)を自然言語で定義することにより、本人工知能装置はその内容を逐次、解釈し必要な処理を進めていくことができる。処理の結果、新しく生成される情報は処理に応じた適切な単語の組合せを検索キーとして記録するので、別の処理において生成した情報を使用する場合でも容易に検索することができる。つまり生成した情報、知識を知識体系に組み込み、将来において必要になった場合に活用することが容易にできるシステムとなっている。本発明では処理の結果生成した新規の情報を逐次、知識体系に組み込み、別の処理において知識体系に組み込まれた情報を適宜、検索して処理を進めていくことができる。また情報の検索対象も知識体系全体、直近に処理した処理結果に限定する等、処理に応じて適切に選択することが可能である。
人間が思考(言語)により解決可能な問題に関しては、その問題を解決する手法または手順を自然言語で示すことにより、本発明の人工知能装置は、入力された問題解決の手法または手順に従って、自律的に問題を解決することができる。
問題解決の手法または手順にあいまいな点、不確定な点があれば、その都度、通知し解決策の手法及び手順の明確化を図りながら問題を解決していくことができる。さらに、人間の問題解決策および行動決定策を自然言語(文)で入力することにより本発明は自律的に条件付処理に変換した上で、条件の成立性を確認しながら処理を進めていくという動作が可能である。人間の問題解決および行動決定に相当する動作をプログラミングすることなく自然言語(文)で表現された知識(問題解決および行動決定に関する手順および思考方法)を入力することにより、人間が思考により問題解決または行動決定するように問題解決または行動決定を実施することが可能となる。
問題解決に必要な記録ユニットを逐次、起動していくことにより問題を解決するための処理および処理プログラムの識別と起動を自律的に実行し問題を解決する。文脈の検出と学習を実施し、文脈を踏まえた会話(質疑、応答、関心のある話題の提供、意見および議論)を実施するとともに関心のある、または指定した相手および事項に関する情報(関心のある話題、問題に対する考え方、物事に対する反応、意見および行動、詳細情報)を収集して記録、蓄積することが可能である。
本発明の知識体系は真実、事実、規則、常識、定義、論理に基づいて構築されている。これらと整合しない情報は知識体系に取り込まない。また全ての処理過程は人間が理解できる言語または記号で表示することが可能な透明性のあるシステムとなっている。人間が蓄積してきた正しい知識を源泉として知識体系を構築し、全ての処理過程における処理状況も人間が理解できる形で表示することが可能なシステムとなっているので「記号接地問題」を解決した「暴走しない人工知能」の実現が可能である。また、問題解決にあたっては構築した知識体系を活用するとともに、問題解決に必要な処理プログラムを自律的に起動し処理結果を知識体系に組み込んで知識体系のさらなる向上を図ることができる。人間が蓄積してきた知識を人工知能装置内に知識体系として構築することができ、必要に応じて人間が開発してきたプログラムを自律的に起動することができるので、問題解決にあたり人間の柔軟性とコンピュータの処理能力を併せ持つ人工知能の実現が可能となる。
According to the present invention, information source, reliability, novelty, field, theme, subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how In addition, it is possible to analyze why it was done, organize it into a sentence structure, and autonomously record information that is judged useful, and construct it as a knowledge system. In addition, the input information is evaluated against the knowledge system built internally, and processing (recording of information, updating / improvement of the knowledge system, execution of instructions, answering questions) is performed autonomously according to the evaluation results. To do. In addition, some of the words included in the information are generalized, and the relationship between the information that is input sequentially and the relationship between the information is enhanced by strengthening the relationship between the set of patterns corresponding to the information and the set of patterns. It is possible to extract relationships and autonomously construct common sense, general ideas and problem solving methods from a series of input information.
Conventionally, when a machine performs processing of input information, etc., it has been necessary to set a program in the computer in advance. It was necessary to create a program for judging the situation from the input information and a program for causing the machine to operate according to individual conditions, and to implement and execute it on a computer mounted on the machine. The program had to be created in a dedicated programming language, and there was a demerit that it took a lot of time for development. If the detection of the condition and the corresponding operation are not appropriate, there is a demerit that it is necessary for a person to correct the program installed in the computer, and it takes a long time for the correction.
According to the present invention, processing is performed by learning how to process information, so that it is not necessary to perform sequential programming. By inputting information on the processing method in a natural language, the machine records and learns the processing method, so that labor can be greatly reduced.
In addition, since the process can be changed by changing the pattern and changing the connection between the patterns regardless of the program, the system is very flexible and adaptable.
In ordinary programming, it is necessary to determine the definition of variables and constants and the contents of processing in advance, so that it takes a lot of labor to realize processing that causes many branches depending on the processing result. (In reality, it is often difficult to realize.) On the other hand, in the present invention, information corresponding to variables and constants used for processing is searched by using a pattern combining natural languages (words and symbols). It is a search method for information that is closer to human thinking. Since it is a system that embodies the recording and retrieval of information in a so-called associative memory system, it is not necessary to strictly define in advance the storage location of information corresponding to variables and constants used for processing. When new information is generated, a combination pattern of natural languages (words and symbols) can be sequentially registered and recorded as a search key. In addition, the registered / recorded information can be easily searched with a search pattern that combines natural languages (words, symbols). This makes the system unconstrained by the constraints of defining variables and constants in programming.
Furthermore, with regard to the content of the processing, the processing corresponding to the condition is performed according to the situation (identifying the validity of the condition based on the pattern and the analysis result of the pattern). If they appear, it is possible to proceed sequentially by analyzing the conditions and determining the processing. Therefore, by defining the same thinking process (a set of conditional sentences) that a human being solves a problem in natural language, the artificial intelligence apparatus can sequentially interpret the contents and proceed with necessary processing. As a result of processing, newly generated information is recorded as a search key with an appropriate combination of words corresponding to the processing, so that it is possible to easily search even when using information generated in another processing. In other words, the generated information and knowledge are incorporated into the knowledge system and can be easily utilized when needed in the future. In the present invention, new information generated as a result of processing can be sequentially incorporated into the knowledge system, and information incorporated into the knowledge system in another process can be appropriately searched and processed. Also, the information search target can be appropriately selected according to the processing, such as limiting the entire knowledge system and the processing result processed most recently.
For problems that humans can solve by thinking (language), the artificial intelligence device of the present invention is autonomous in accordance with the input problem solving technique or procedure by indicating the technique or procedure for solving the problem in natural language. Problem can be solved.
If there is an ambiguity or uncertain point in the problem solving technique or procedure, the problem can be solved by notifying each time and clarifying the solution technique and procedure. Furthermore, by inputting human problem solving measures and action decision measures in natural language (sentences), the present invention autonomously converts them into conditional processing, and then proceeds with the processing while confirming the establishment of the conditions. Is possible. By inputting knowledge (procedures and methods for problem solving and action determination) expressed in natural language (sentences) without programming the actions equivalent to human problem solving and action determination, humans can solve problems through thinking. Alternatively, problem solving or action determination can be performed so as to determine action.
By sequentially starting the recording units necessary for problem solving, the process for solving the problem and the identification and activation of the processing program are autonomously executed to solve the problem. Detect and learn context, conduct contextual conversations (questions, responses, presentation of topics of interest, opinions and discussions) and information about interested or specified parties and matters (interested) It is possible to collect, record, and accumulate topics, ideas about problems, responses to things, opinions and actions, and detailed information.
The knowledge system of the present invention is constructed based on truth, facts, rules, common sense, definition and logic. Information that does not match these is not taken into the knowledge system. All processing steps are transparent systems that can be displayed in human-understood language or symbols. The knowledge system has been constructed using the correct knowledge accumulated by humans as the source, and the system can display the processing status in all processing processes in a form that can be understood by humans. Realization of “artificial intelligence that does not run away” is possible. Further, in solving problems, the constructed knowledge system can be utilized, and a processing program necessary for problem solving can be activated autonomously, and the processing results can be incorporated into the knowledge system to further improve the knowledge system. Knowledge accumulated by human beings can be constructed as a knowledge system in an artificial intelligence device, and programs developed by human beings can be launched autonomously as needed. It is possible to realize artificial intelligence that has both computer processing capabilities.

人工知能装置の構成例Configuration example of artificial intelligence device 語列の励起履歴の例Example of excitation history of word sequence 文構造を分析する機能例Example of analyzing sentence structure 文構造の分析例Example of sentence structure analysis 記録した情報から質問に対する回答を生成する機能例Example function to generate an answer to a question from recorded information 情報をパターンに変換しパターン記録器に記録する動作例Example of operation that converts information into a pattern and records it in a pattern recorder 入力した情報と記録した情報の比較動作例Example of comparison between input information and recorded information 情報を評価して処置(記録、破棄)の実施例Example of measures (recording, discarding) by evaluating information 励起したパターンの履歴をパターン照射器に記録する動作例Example of recording the history of the excited pattern in the pattern irradiator 記録ユニットの機能概要Overview of recording unit functions 問題を検出して自律的に解決する方法の機能例Functional examples of how to detect problems and solve them autonomously 問題を検出して自律的に解決するパターン遷移例Pattern transition example to detect problems and solve them autonomously

1 パターン変換器
2 パターン記録器
3 パターン分析器
4 パターン制御器
5 パターン逆変換器
6 パターン照射器
1 Pattern Converter 2 Pattern Recorder 3 Pattern Analyzer 4 Pattern Controller 5 Pattern Reverse Converter 6 Pattern Irradiator

Claims (12)

情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した情報と既に記録している情報との関係を分析し有用な情報を抽出して知識体系の構築、拡大および更新を実施し、知識体系に記録された情報、情報間の関係および処理プログラムにより生成した情報を使用して問題解決する人工知能装置。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns Pattern recorder for recording pattern analysis results, and pattern controller for registering, starting, changing, processing, controlling and starting processing programs according to human instructions, pattern states and pattern analysis results A pattern irradiator that records the pattern activation history, analyzes the situation, context and problem using the recorded activation history information, and activates the pattern and processing program necessary for problem solving; and A pattern inverse converter that converts information into information, analyzes the relationship between the input information and the information already recorded, extracts useful information, constructs, expands, and updates the knowledge system. An artificial intelligence device that solves problems using recorded information, relationships between information, and information generated by a processing program. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、人間が入力する情報から知識体系を構築し、人間との話し相手、相談相手およびアシスタントとして動作する人工知能装置。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations Relationship with patterns Pattern recorder for recording pattern analysis results, and pattern controller for registering, starting, changing, processing, controlling and starting processing programs according to human instructions, pattern states and pattern analysis results A pattern irradiator that records the pattern activation history, analyzes the situation, context and problem using the recorded activation history information, and activates the pattern and processing program necessary for problem solving; and Artificial intelligence device that has a pattern inverse converter that converts information into information, constructs a knowledge system from information input by humans, and operates as a talker, consultant, and assistant with humans. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した真実、事実、常識、規則、専門知識から知識体系を構築し人間を支援する人工知能装置。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns Pattern recorder for recording pattern analysis results, and pattern controller for registering, starting, changing, processing, controlling and starting processing programs according to human instructions, pattern states and pattern analysis results A pattern irradiator that records the pattern activation history, analyzes the situation, context and problem using the recorded activation history information, and activates the pattern and processing program necessary for problem solving; and Artificial intelligence device that supports humans by building a knowledge system from input truth, facts, common sense, rules, and expertise, with a pattern inverse converter that converts information into information. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した真実、事実、常識、規則、専門知識から知識体系を構築した1つ以上の専門分野用人工知能装置と専門分野用人工知能装置を連携させるマネージメント用人工知能装置から構成され、解決に複数の専門知識が必要な問題の解決を支援する人工知能装置。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns Pattern recorder for recording pattern analysis results, and pattern controller for registering, starting, changing, processing, controlling and starting processing programs according to human instructions, pattern states and pattern analysis results A pattern irradiator that records the pattern activation history, analyzes the situation, context and problem using the recorded activation history information, and activates the pattern and processing program necessary for problem solving; and A pattern inverse converter that converts information into information, and links one or more artificial intelligence devices that have built up a knowledge system from input truths, facts, common sense, rules, and specialized knowledge with artificial intelligence devices for specialized fields An artificial intelligence device that consists of artificial intelligence devices for management and supports the solution of problems that require multiple specialized knowledge to solve. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御、パターン間の関連付けおよび処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した情報と既に記録している情報との関係を分析し有用な情報を抽出して知識体系の構築、拡大および更新を実施し、知識体系に記録された情報、情報間の関係および処理プログラムにより生成した情報を使用して問題解決する人工知能装置。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns A pattern recorder that records pattern analysis results and pattern registration, activation, modification, processing, control, pattern correlation, and processing program activation according to human instructions, pattern status, and pattern analysis results A pattern controller that records the activation history of the pattern, the pattern activation history, uses the recorded activation history information to analyze the situation, context, and problem, and activates the pattern and processing program required to solve the problem Equipped with an irradiator and a pattern inverse converter that converts patterns into information, analyzes the relationship between the input information and the information already recorded, extracts useful information, and builds, expands and updates the knowledge system An artificial intelligence device that solves problems using information recorded in the knowledge system, relationships between information, and information generated by a processing program. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御、パターン間の関連付けおよび処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した情報と既に記録している情報との関係を分析し有用な情報を抽出して知識体系の構築、拡大および更新を実施し、知識体系に記録された情報、情報間の関係および処理プログラムにより生成した情報を使用して会話する人工知能装置。A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations Relationship with patterns A pattern recorder that records pattern analysis results and pattern registration, activation, modification, processing, control, pattern correlation, and processing program activation according to human instructions, pattern status, and pattern analysis results A pattern controller that records the activation history of the pattern, the pattern activation history, uses the recorded activation history information to analyze the situation, context, and problem, and activates the pattern and processing program required to solve the problem Equipped with an irradiator and a pattern inverse converter that converts patterns into information, analyzes the relationship between the input information and the information already recorded, extracts useful information, and builds, expands and updates the knowledge system And an artificial intelligence apparatus that uses information recorded in the knowledge system, the relationship between the information, and information generated by a processing program for conversation. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した情報と既に記録している情報との関係を分析し有用な情報を抽出して知識体系の構築、拡大および更新を実施し、知識体系に記録された情報、情報間の関係および処理プログラムにより生成した情報を使用して問題解決する人工知能装置として機能させるためのソフトウェア。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns Pattern recorder for recording pattern analysis results, and pattern controller for registering, starting, changing, processing, controlling and starting processing programs according to human instructions, pattern states and pattern analysis results A pattern irradiator that records the pattern activation history, analyzes the situation, context and problem using the recorded activation history information, and activates the pattern and processing program necessary for problem solving; and A pattern inverse converter that converts information into information, analyzes the relationship between the input information and the information already recorded, extracts useful information, constructs, expands, and updates the knowledge system. Software to function as an artificial intelligence device that solves problems using recorded information, information relationships, and information generated by processing programs Towea. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、人間が入力する情報から知識体系を構築し、人間との話し相手、相談相手およびアシスタントとして動作する人工知能装置として機能させるためのソフトウェア。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations Relationship with patterns Pattern recorder for recording pattern analysis results, and pattern controller for registering, starting, changing, processing, controlling and starting processing programs according to human instructions, pattern states and pattern analysis results A pattern irradiator that records the pattern activation history, analyzes the situation, context and problem using the recorded activation history information, and activates the pattern and processing program necessary for problem solving; and Software that includes a pattern inverse converter that converts information into information, constructs a knowledge system from information input by humans, and functions as an artificial intelligence device that operates as a person to talk to, a person to consult with, and an assistant. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した真実、事実、常識、規則、専門知識から知識体系を構築し人間を支援する人工知能装置として機能させるためのソフトウェア。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns Pattern recorder for recording pattern analysis results, and pattern controller for registering, starting, changing, processing, controlling and starting processing programs according to human instructions, pattern states and pattern analysis results A pattern irradiator that records the pattern activation history, analyzes the situation, context and problem using the recorded activation history information, and activates the pattern and processing program necessary for problem solving; and Is a software that has a pattern inverse converter that converts information into information and functions as an artificial intelligence device that supports human beings by constructing a knowledge system from input truths, facts, common sense, rules, and expertise. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御および処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した真実、事実、常識、規則、専門知識から知識体系を構築した1つ以上の専門分野用人工知能装置と専門分野用人工知能装置を連携させるマネージメント用人工知能装置から構成され、解決に複数の専門知識が必要な問題の解決を支援する人工知能装置として機能させるためのソフトウェア。  A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns Pattern recorder for recording pattern analysis results, and pattern controller for registering, starting, changing, processing, controlling and starting processing programs according to human instructions, pattern states and pattern analysis results A pattern irradiator that records the pattern activation history, analyzes the situation, context and problem using the recorded activation history information, and activates the pattern and processing program necessary for problem solving; and A pattern inverse converter that converts information into information, and links one or more artificial intelligence devices that have built up a knowledge system from input truths, facts, common sense, rules, and specialized knowledge with artificial intelligence devices for specialized fields Consists of artificial intelligence devices for management and functions as an artificial intelligence device that helps solve problems that require multiple specialized knowledge to solve. Because of the software. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御、パターン間の関連付けおよび処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した情報と既に記録している情報との関係を分析し有用な情報を抽出して知識体系の構築、拡大および更新を実施し、知識体系に記録された情報、情報間の関係および処理プログラムにより生成した情報を使用して問題解決する人工知能装置として機能させるためのソフトウェア。A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns A pattern recorder that records pattern analysis results and pattern registration, activation, modification, processing, control, pattern correlation, and processing program activation according to human instructions, pattern status, and pattern analysis results A pattern controller that records the activation history of the pattern, the pattern activation history, uses the recorded activation history information to analyze the situation, context, and problem, and activates the pattern and processing program required to solve the problem Equipped with an irradiator and a pattern inverse converter that converts patterns into information, analyzes the relationship between the input information and the information already recorded, extracts useful information, and builds, expands and updates the knowledge system An artificial intelligence device that solves problems using information recorded in the knowledge system, relationships between information, and information generated by processing programs; Software to function Te. 情報(単語、言語、文章、知識、数式、記号、画像、音声)をパターンに変換するパターン変換器と、入力した情報に関して情報の源泉、信頼性、新規性、分野、テーマ、話題、意図、主語、主語の修飾、述語、述語の修飾、修飾関係、いつ、どこで、誰が、何を、どのように、なぜ、したのか、文の種類(平常文、疑問文、命令文、条件文、真実、事実、規則、常識、善悪、定義、論理、説明、仮説、予測、意見、感想、噂)、数式、物理式、化学式、記号および記録している関連情報との関係を分析し、単語、文要素、文および分析結果を使用した検索ができる構造に整理してパターン記録器に記録するパターン分析器と、パターン、パターンおよび処理プログラムの起動の条件、パターン間の接続関係、接続関係のあるパターンとの関係およびパターンの分析結果を記録するパターン記録器と、人間の指示、パターンの状態およびパターンの分析結果に応じてパターンの登録、起動、変更、処理、制御、パターン間の関連付けおよび処理プログラムの起動を実施するパターン制御器と、パターンの起動の履歴を記録し、記録した起動履歴の情報を使用して状況、文脈および問題の分析を実施し、問題解決に必要なパターンおよび処理プログラムを起動するパターン照射器と、パターンを情報に変換するパターン逆変換器を備え、入力した情報と既に記録している情報との関係を分析し有用な情報を抽出して知識体系の構築、拡大および更新を実施し、知識体系に記録された情報、情報間の関係および処理プログラムにより生成した情報を使用して会話する人工知能装置として機能させるためのソフトウェア。A pattern converter that converts information (words, languages, sentences, knowledge, mathematical formulas, symbols, images, sounds) into patterns, and information sources, reliability, novelty, fields, themes, topics, intentions, Subject, subject modification, predicate, predicate modification, modification relationship, when, where, who, what, how, why, type of sentence (normal sentence, question sentence, imperative sentence, conditional sentence, truth Analyze facts, rules, common sense, good and evil, definitions, logic, explanations, hypotheses, predictions, opinions, impressions, rumors), formulas, physical formulas, chemical formulas, symbols and related information that you record, words, There is a pattern analyzer that organizes into a structure that can be searched using sentence elements, sentences, and analysis results, and records them in the pattern recorder, and the pattern, pattern and processing program start conditions, connection relations between patterns, and connection relations. Relationship with patterns A pattern recorder that records pattern analysis results and pattern registration, activation, modification, processing, control, pattern correlation, and processing program activation according to human instructions, pattern status, and pattern analysis results A pattern controller that records the activation history of the pattern, the pattern activation history, uses the recorded activation history information to analyze the situation, context, and problem, and activates the pattern and processing program required to solve the problem Equipped with an irradiator and a pattern inverse converter that converts patterns into information, analyzes the relationship between the input information and the information already recorded, extracts useful information, and builds, expands and updates the knowledge system As an artificial intelligence device that uses the information recorded in the knowledge system, the relationship between the information, and information generated by the processing program to converse Software to function.
JP2015205904A 2015-09-30 2015-09-30 Artificial intelligence device for expanding processing capacity in self-organizing manner Pending JP2016028346A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2015205904A JP2016028346A (en) 2015-09-30 2015-09-30 Artificial intelligence device for expanding processing capacity in self-organizing manner

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2015205904A JP2016028346A (en) 2015-09-30 2015-09-30 Artificial intelligence device for expanding processing capacity in self-organizing manner

Publications (1)

Publication Number Publication Date
JP2016028346A true JP2016028346A (en) 2016-02-25

Family

ID=55360761

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2015205904A Pending JP2016028346A (en) 2015-09-30 2015-09-30 Artificial intelligence device for expanding processing capacity in self-organizing manner

Country Status (1)

Country Link
JP (1) JP2016028346A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2020161111A (en) * 2019-03-27 2020-10-01 ワールド ヴァーテックス カンパニー リミテッド Method for providing prediction service of mathematical problem concept type using neural machine translation and math corpus
CN114898751A (en) * 2022-06-15 2022-08-12 中国电信股份有限公司 Automatic configuration method and system, storage medium and electronic equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015133138A (en) * 2015-02-17 2015-07-23 洋彰 宮崎 autonomous intelligence generator
JP2015144009A (en) * 2015-03-03 2015-08-06 洋彰 宮崎 Autonomous intelligence scheme analysis construction machine
JP2015164064A (en) * 2015-04-09 2015-09-10 洋彰 宮崎 Autonomous knowledge extracting device
JP2015165414A (en) * 2015-03-27 2015-09-17 洋彰 宮崎 Autonomous type problem solving machine

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015133138A (en) * 2015-02-17 2015-07-23 洋彰 宮崎 autonomous intelligence generator
JP2015144009A (en) * 2015-03-03 2015-08-06 洋彰 宮崎 Autonomous intelligence scheme analysis construction machine
JP2015165414A (en) * 2015-03-27 2015-09-17 洋彰 宮崎 Autonomous type problem solving machine
JP2015164064A (en) * 2015-04-09 2015-09-10 洋彰 宮崎 Autonomous knowledge extracting device

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2020161111A (en) * 2019-03-27 2020-10-01 ワールド ヴァーテックス カンパニー リミテッド Method for providing prediction service of mathematical problem concept type using neural machine translation and math corpus
CN114898751A (en) * 2022-06-15 2022-08-12 中国电信股份有限公司 Automatic configuration method and system, storage medium and electronic equipment
CN114898751B (en) * 2022-06-15 2024-04-23 中国电信股份有限公司 Automatic configuration method and system, storage medium and electronic equipment

Similar Documents

Publication Publication Date Title
JP5907469B2 (en) Artificial intelligence device that autonomously expands knowledge by language input
Chowdhary Fundamentals of artificial intelligence
Riley et al. Integrating non-monotonic logical reasoning and inductive learning with deep learning for explainable visual question answering
Averkin et al. Review of research in the field of developing methods to extract rules from artificial neural networks
Crowder et al. Artificial Psychology
Amnueypornsakul et al. Machine-guided solution to mathematical word problems
Chowdhary et al. Introducing artificial intelligence
JP5881030B2 (en) Artificial intelligence device that expands knowledge in a self-organizing manner
Kostopoulos et al. Deep dense neural network for early prediction of failure-prone students
JP5910957B2 (en) Artificial intelligence device that autonomously constructs a knowledge system by language input
JP2017208054A (en) Artificial intelligence device for searching information and procedure by combining concept, and for autonomously processing problem
JP2016028346A (en) Artificial intelligence device for expanding processing capacity in self-organizing manner
JP5854251B2 (en) Artificial intelligence device
JP2017220191A (en) Artificial intelligence carrying out autonomous processing of problem by combination of procedures
JP2017130171A (en) Non-black box artificial intelligence device
JP5854392B2 (en) Artificial intelligence device that does not run away
JP5854250B2 (en) Artificial intelligence device
JP5828438B2 (en) Autonomous knowledge system construction machine
Habitamu Designing and Implementing Adaptive Bot Model to Consult Ethiopian Published Laws Using Ensemble Architecture with Rules Integrated
Sohn et al. Evaluating comprehension-based user models: Predicting individual user planning and action
Zhang et al. Inducing grammar from long short-term memory networks by shapley decomposition
Strømsvåg Exploring the Why in AI: Investigating how Visual Question Answering models can be interpreted by post-hoc linguistic and visual explanations
Drigas et al. Decade review (1999-2009): progress of application of artificial intelligence tools in student diagnosis
Gibaut et al. Neurosymbolic AI and its Taxonomy: a survey
Ghelani A review on student automatic grading system

Legal Events

Date Code Title Description
A975 Report on accelerated examination

Free format text: JAPANESE INTERMEDIATE CODE: A971005

Effective date: 20151222

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20160112

A02 Decision of refusal

Free format text: JAPANESE INTERMEDIATE CODE: A02

Effective date: 20160517