JP2015165414A - Autonomous type problem solving machine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate an answer by mutually relating and constructing information as a knowledge system by person's instruction or learning, and using a solving method which is taught autonomously in the past with respect to a given problem.SOLUTION: A machine is made to learn methods of processing input information such as a method of analyzing relation between information and information, a method of constructing a knowledge system, and a method of solving a problem, and the machine analyzes the input information as to its subject, predicate, modification of the subject, modification of the predicate, modification relation, when, where, who, what, and why, interprets and records meaning of the information, and constructs truth, reality, fact, technical knowledge, rule, and common sense as a knowledge system. The machine records and learns methods taught by a person in the beginning, and autonomously executes processing after having learnt to some extent, and the person confirms a processing result of the machine, and informs the machine that the processing is wrong in such a case so as to properly correct the processing. In the learning phase, a series of patterns corresponding to the process are sequentially excited as instructed by the person and the history thereof is recorded.

Description

  The present invention relates to input information, analyzes subject, predicate, subject modification, predicate modification, modification relationship, when, where, who, what, how and why, and interprets the meaning of the information Analyzing the relationship between information and information, and building truth, truth, facts, expertise, rules and common sense as a knowledge system. It also relates to artificial intelligence and software that uses the constructed knowledge system to autonomously generate answers to questions, process instructed by language, and solutions to given problems.

  Conventional artificial intelligence performs information processing according to a preprogrammed procedure. Implementation other than the processing set by the program is difficult. Moreover, it has been difficult to construct a knowledge system by associating information with each other by analyzing the meaning of input information and analyzing the relationship with information recorded in the past by human instruction or learning. In addition, it is difficult to generate an answer using a solution autonomously taught in the past for a given problem.

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 install 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. Moreover, it has been difficult to construct a knowledge system by associating information with each other by analyzing the meaning of input information and analyzing the relationship with information recorded in the past by human instruction or learning. In addition, it is difficult to generate an answer using a solution autonomously taught in the past for a given problem.
In the present invention, processing for analyzing and recording information and the structure of information by human instruction and learning is performed, and the relationship between information is constructed as a knowledge system by connection between patterns and processing between patterns. Analyze the subject, predicate, subject modification, predicate modification, modification relationship, when, where, who, what, how, why, and interpret the meaning of the information. At the same time, the relationship between information and information is analyzed, and truth, truth, facts, expertise, rules and common sense are constructed as a knowledge system. In addition, using the constructed knowledge system, it automatically generates answers to questions, processes instructed by language, and solutions to given problems.

When the machine performs processing of input information, etc., it is realized by incorporating and executing a program created in advance in a program language in a computer mounted on the machine. A program is created such that when a preset condition is detected, a corresponding operation is executed. If the detection of the condition and the corresponding action are not appropriate, the program installed on the computer is modified.
Analyze the subject, predicate, subject modification, predicate modification, modification relationship, when, where, who, what, how, why, and interpret the meaning of the information. At the same time, truth, truth, facts, expertise, rules, and common sense are constructed as a knowledge system, and the constructed knowledge system is used to generate answers to questions, process directed by language, and solutions to given problems. There has been no artificial intelligence and software that automatically generates.

Problems to be solved by the invention

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 install 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.
It also analyzes the subject, predicate, subject modification, predicate modification, modification relationship, when, where, who, what, how, why, and the meaning of the information. Record and build truth, truth, facts, expertise, rules, and common sense as a knowledge system, then use the built knowledge system to generate answers to questions, process directed by language, and given problems It was difficult to generate a solution autonomously.

Means for solving the problem

In the present invention, the input information processing is not individually programmed, but the machine learns the input information processing method such as a method of analyzing the relationship between information and information, a knowledge system construction method, and a problem solving method. To implement.
Analyzing the subject information, subject, predicate, subject modification, predicate modification, modification relationship, when, where, who, what, how, why, and interpreting and recording the meaning of the information , Build truth, truth, facts, expertise, rules and common sense as a knowledge system.
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.
Human thought is expressed in language, but the information expressed in this language is converted into what is called a pattern. When converted into a pattern, individual human thoughts can be expressed as individual patterns, and human thought transitions can be viewed as patterns to patterns.
The pattern is not only a language expression but also a concept such as a sentence and a sentence. In addition, it is possible to sequentially excite related patterns and execute a number of processes on the excited patterns. Furthermore, it is a concept that has a very wide range that can be handled, for example, it is possible to generate a signal for driving the driving device in order to perform processing and operation of image information and information.

When we convert the language that expresses human thoughts into patterns and analyze the patterns and the relationships between patterns, the types of sentences (normal sentences, question sentences, imperative sentences, etc.) and features (truths, truths, facts, definitions, rules, Common sense, explanation, hypothesis, prediction, opinion, impression).
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, etc. by analyzing words included in the sentence. Among the features of the sentence, truth, truth, facts, definitions, rules, and common sense are taught by humans for each individual and learned by the machine. This can be done by identifying and recording each type of information (truth, truth, fact, definition, rule, common sense) as each information is entered into 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 is word identification from word sequence, word attributes, sentence element analysis, sentence structure analysis (subject, predicate, subject modification, predicate modification, modification relation), relation analysis between sentence element and sentence element (Same meaning, definition, opposite meaning, etc.) are implemented, and the relationship between the sentence element and the sentence element is associated as the connection relation between the pattern and the pattern, thereby converting the relation between the information and the meaning of the information into the connection relation between the pattern and the pattern. I will do it. The relationship between patterns can represent not only sentence elements and sentence elements but also 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. The relationship between the recorded patterns can be utilized when controlling the transition 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.
In this way, by defining the relationship between information as a connection relationship between patterns, information is not recorded in a single state, but as a knowledge system with 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.

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 related to the contents even if the words do not match.
The newly entered information can be checked against the already recorded knowledge system and evaluated for consistency and novelty. Information that is identified as truth, truth, facts, definitions, rules and common sense and is not consistent with the information recorded is likely to be incorrect. On the other hand, if it is not possible to judge whether the information is consistent or not, it is considered that knowledge sufficient to determine the information has not yet been stored, and therefore human judgment is required. If humans can judge that the information is correct, the result will be notified to the machine. Information that has been evaluated when the machine can logically develop the combination of truth, truth, facts, definitions, rules, and common sense by recording the information as an explanation or hypothesis and obtaining other information in the future. Will be identified and recorded.
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.

When information is constructed as a knowledge system in this way, problem solving methods can be learned and knowledge can be improved. First, analyze the problem, clarify the problem, and set a goal for problem solving. 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.
Problem analysis and problem clarification are implemented by learning how to detect and identify differences between the target state and the current situation. Moreover, the solution proposal corresponding to each problem is implemented by learning the corresponding processing.
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 generally be expressed as conditional processing. In the present invention, the meaning of the language is interpreted and autonomously converted as conditional processing. The condition of the conditional processing is determined by generating a search pattern from the corresponding word and searching 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 the corresponding information is acquired, the process according to the content is performed.
In the learning phase, when a pattern is excited, the history of the excited pattern is referred to during the set period before the pattern is excited, and the connection relationship with the excited pattern is strengthened. In the practical phase, the history of the excitation pattern in the set period is collated with the record of the connection relationship with other patterns recorded for each recording module, and the pattern is excited when the correlation is large. When the pattern is excited, the history of the excitation pattern is updated, and patterns having a large correlation with the history are sequentially excited in the new state.

In this way, in the learning phase, a series of patterns corresponding to processing is sequentially excited by a human instruction, and by recording the history, the corresponding pattern is excited according to the instructed procedure. The behavior of the pattern is not static but shows dynamic behavior. Using the internal pattern recorded in the pattern, it is possible to search for necessary information and store the search result in a required place. Also, it is possible to perform processing such as converting the internal pattern recorded in the pattern into a designated arrangement. Numerous processes can be implemented by combining patterns that can behave in such a dynamic manner.
Teaching to the autonomous problem-solving machine can be implemented by sequentially inputting language information without programming. The input language information is analyzed for syntax, meaning, and relationship with already recorded information, and the corresponding pattern is excited according to the analysis result, and processing is executed. As specific processing, various processing such as value evaluation and recording of input information, execution of an instructed command, generation of a solution to a problem / issue can be performed.
The overall operation of this autonomous problem solver 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.).

FIG. 1 is a diagram showing the configuration of an autonomous problem solving machine in one embodiment of the present invention.
In FIG. 1, reference numeral 1 denotes a pattern converter that converts information into a pattern. Reference numeral 2 denotes a pattern recorder that records patterns, connection relationships between patterns, and relationships between patterns. Reference numeral 3 denotes a pattern register that registers and changes a pattern and a connection relation between the patterns by a human instruction or autonomously. Reference numeral 4 denotes a pattern controller for controlling pattern processing. Reference numeral 5 denotes a pattern inverse converter for converting a pattern into information. Reference numeral 6 denotes a pattern analyzer that analyzes patterns and relationships between patterns. 7 records the excitation history of the pattern, and when the pattern is excited, generates connection relationship data with the pattern from the history data of the pattern excited before that, and the connection relationship recording unit of the recording module of the pattern A pattern illuminator that excites a recording module having a high correlation by comparing the history data of the excitation pattern from the present to the past and the connection relationship data recorded in the connection relationship recording section of each recording module. is there.

Next, the operation will be described.
In FIG. 1, a pattern converter 1 converts information into a pattern. The converted pattern is analyzed for the meaning of the pattern, and processing according to the analysis result is performed in the pattern controller 4. The pattern recorder 2 records patterns, connection relationships between patterns, and relationships between patterns. The pattern register 3 performs pattern registration and change. The input pattern is collated with the recording module of the pattern recorder, and it is confirmed whether or not the related pattern is recorded. If the same or equivalent pattern as the input pattern is recorded, the corresponding pattern is excited, and if not recorded, it is registered as a new pattern and excited. The history of the excited pattern is recorded in 7 pattern irradiators. When a recording module is excited, it generates connection relationship data with the pattern from the history data of the pattern excited before that, and records it in the connection relationship recording section of the recording module for the pattern. The excitation pattern history data and the connection relation data recorded in the connection relation recording section of each recording module are collated to excite a recording module having a large correlation. In the initial stage, patterns and patterns are generated by teaching from humans.

In the present invention, the input information processing is not individually programmed, but the machine learns the input information processing method such as a method of analyzing the relationship between information and information, a knowledge system construction method, and a problem solving method. To implement.
Analyzing the subject information, subject, predicate, subject modification, predicate modification, modification relationship, when, where, who, what, how, why, and interpreting and recording the meaning of the information , Build truth, truth, facts, expertise, rules and common sense as a knowledge system.
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.
Human thought is expressed in language, but the information expressed in this language is converted into what is called a pattern. When converted into a pattern, individual human thoughts can be expressed as individual patterns, and human thought transitions can be viewed as patterns to patterns.
The pattern is not only a language expression but also a concept such as a sentence and a sentence. In addition, it is possible to sequentially excite related patterns and execute a number of processes on the excited patterns. Furthermore, it is a concept that has a very wide range that can be handled, for example, it is possible to generate a signal for driving the driving device in order to perform processing and operation of image information and information.

FIG. 2 shows an example of a pattern. The input sentence is organized into a main part (subject and subject modification) and a predicate (predicate and predicate modification). The predicate modification further organizes what, when, where, why, and how. Organizing and storing in this way is very useful for retrieving information. The input word pattern is stored as an in-sentence pattern. Relationships between words are analyzed, sentence elements (subjects, predicates, modifiers) and modifier relationships are analyzed.
FIG. 3 shows an operation in which words, sentence elements, and modification relationships are sequentially identified from a word string. When a word is input, the recording module corresponding to the input word is excited in the false detection area. When the recording module corresponding to the word is sequentially excited, when the word corresponding to the word string is detected, the recording module corresponding to the word is excited. The connection relation recording unit of the recording module corresponding to each word generates a connection relation with the word string corresponding to each word, so that it is detected that the correlation is large when the history of the pattern irradiator is irradiated. Then, the recording module is excited. When the word pattern is excited, the part-of-speech / meaning pattern of the word corresponding to the word is excited. Next, the sentence element and the modification relationship between the sentence elements are detected according to the appearance pattern of the word part of speech. If there are multiple words to be modified, they are located apart from the word to be modified, and it is difficult to identify which word is modifying which word by order of part of speech, the identification is also performed using the meaning of the word To do. (This can be done by detecting the feasibility of the combination of the word to be modified and the word to be modified.) When the relationship between the sentence elements (subject, predicate, modification relationship) is identified, the relationship between the sentence elements Is recorded. Sentences input from this relationship are organized into a main part (subject and subject modification) and predicate (predicate and predicate modification). The predicate modification further includes what, when, where, why, and how. Are stored in the recording module as in-sentence patterns.

Next, how the input sentence is analyzed and recorded as an in-sentence pattern in the recording module will be described.
FIG. 4 shows an example of current affairs. The procedure for analyzing the text example by the processing of FIG. 3 is shown in FIGS. When a word is detected, the part of speech and meaning of the word are analyzed at the same time. Nouns, verbs, adjectives, adjective verbs, adverbs, particles, etc. are identified as parts of speech. Subject identification and modifier identification are performed based on the type of noun and particle. If there are multiple words to be modified, they are located apart from the word to be modified, and it is difficult to identify which word is modifying which word by order of part of speech, the identification is also performed using the meaning of the word To do. This can be done by detecting the feasibility of the combination of the word to be modified and the word to be modified. By tracing the modification relationship sequentially, the modification of the subject and the modification part of the predicate can be identified. What is (O), when (H1), where (H2), why (H3), and how (H4) is implemented as a modifier of the predicate. This can be implemented by identifying the particle of each sentence element, and identifying whether it is in the middle of a modification relationship or a delimiter between sentence elements of each modification. Through the analysis described above, subject modification, subject, predicate modifier (what, when, where, why, how) and predicates can be identified. This analysis result is defined as the connection relation of the in-sentence pattern, and is recorded as the in-sentence (word) pattern connection information in FIG. By performing such an analysis, it is possible to strictly interpret a sentence that is a compound sentence (a sentence in which a sentence exists and a word is modified). There are usually multiple subjects, predicates, and modifiers in a sentence, but it is important to identify exactly which subject and predicate are the main and what others are modifying. it can. By performing this analysis on the input sentence, the relation between sentences can be strictly identified. Various comparisons (formal comparison, semantic comparison, designation of comparison location, etc.) can be made by comparing sentence subjects, predicates, and modifiers. Also, specify what information (what, who, what, when, where, why, how, and what you want to search) when searching for information from sentences recorded in the past. You can search for the information you want directly.

Next, a method for searching for desired information will be described.
FIG. 10 shows an example of questions related to current affairs cited as an example earlier. FIG. 16 shows an operation for generating an answer to a question. Convert the question sentence to subject, subject modification, predicate, predicate modification (what, when, where, why, how) as described above
I will place a turn. From this, a search pattern corresponding to the question can be generated. Also, what are the expected sentence elements to answer questions (subjects, subject modifiers, predicates,
If there is an internal pattern at the corresponding position, it becomes a candidate for an answer. In the case of compound sentences, the hierarchy of predicates corresponding to the questions is important. In other words, it is important to extract the answer from the hierarchy in which the predicate corresponding to the question is excited when the search pattern is irradiated and the related pattern is searched. When multiple predicates exist in a compound sentence, it is necessary to identify which predicate is expected to be answered. 11 to 15 show where the answers corresponding to the question of FIG. 10 are extracted.

When we convert the language that expresses human thoughts into patterns and analyze the patterns and the relationships between patterns, the types of sentences (normal sentences, question sentences, imperative sentences, etc.) and features (truths, truths, facts, definitions, rules, Common sense, explanation, hypothesis, prediction, opinion, impression, etc.).
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, etc. by analyzing words included in the sentence. Among the features of the sentence, truth, truth, facts, definitions, rules, and common sense are taught by humans for each individual and learned by the machine. This can be done by identifying and recording each type of information (truth, truth, fact, definition, rule, common sense) as each information is entered into the machine. As shown in FIG. 17, the type of information is also recorded in the recording module, so that when the pattern is excited, it is possible to identify what type of information the pattern is. 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 is word identification from word sequence, word attributes, sentence element analysis, sentence structure analysis (subject, predicate, subject modification, predicate modification, modification relation), relation analysis between sentence element and sentence element (Same meaning, definition, opposite meaning, etc.) are implemented, and the relationship between the sentence element and the sentence element is associated as the connection relation between the pattern and the pattern, thereby converting the relation between the information and the meaning of the information into the connection relation between the pattern and the pattern. I will do it. The relationship between patterns can represent not only sentence elements and sentence elements but also 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. The relationship between the recorded patterns can be utilized when controlling the transition 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.
In this way, by defining the relationship between information as a connection relationship between patterns, information is not recorded in a single state, but as a knowledge system with 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.
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 related to the contents even if the words do not match.
The newly entered information can be checked against the already recorded knowledge system and evaluated for consistency and novelty. Information that is identified as truth, truth, facts, definitions, rules and common sense and is not consistent with the information recorded is likely to be incorrect. On the other hand, if it is not possible to judge whether the information is consistent or not, it is considered that knowledge sufficient to determine the information has not yet been stored, and therefore human judgment is required. If humans can judge that the information is correct, the result will be notified to the machine. Information that has been evaluated when the machine can logically develop the combination of truth, truth, facts, definitions, rules, and common sense by recording the information as an explanation or hypothesis and obtaining other information in the future. Will be identified and recorded.
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.

17 to 21 show examples of utilization of the knowledge system. In FIG. 17, if the pattern of the input information is [PA], the pattern of [PA] is irradiated to the recording area storing the meaning of the word and the meaning of the corresponding word is searched. By performing this search in association with the sentence structure (subject, predicate, subject modification, predicate modification), a pattern that is semantically equivalent to the input information can be generated. This semantically equivalent pattern is expressed as [PA #]. When retrieving related information from input information, a number of retrieval methods can be used. Information that exactly matches the input information, information that is semantically equivalent, information that partially matches, etc. can be selected depending on the purpose of the search.
FIG. 18 shows that a semantically equivalent pattern [PA #] is irradiated and [PB] is detected as related information. Similarly, semantic analysis is performed on [PB] to generate a semantically equivalent pattern [PB #].
FIG. 19 shows an analysis of the relationship between input information and related information already recorded. A matching pattern or a different pattern can be extracted from the difference between [PA #] and [PB #]. Further, it is possible to detect a semantic difference by irradiating a recording area in which the relationship between words is defined (equivalent, similar, opposite, etc.) with [PA #] and [PB #]. By making the irradiation lines of the input information and the related information different lines, it is possible to identify whether the excitation of the pattern corresponding to each word is due to the input information or the related information. In the recording area where the relationship between the words is defined, the relationship between the words is identified by the connection relationship. For this reason, if the word excited by the input information and the word excited by the related information are semantically the same, the pattern indicating the same meaning is excited and can be identified as having the same meaning. On the other hand, if the word whose input information is excited and the word whose related information is excited are semantically opposite, the pattern indicating the opposite meaning is excited, and it can be identified that the meaning is opposite. FIG. 21 shows an operation example for detecting a difference in meaning.

FIG. 20 shows an operation example in which input information is compared with related information already recorded, and related information is evaluated. Since the type of related information (truth, truth, fact, expertise, rule, common sense) can be identified, the validity of the input information can be evaluated by comparing the consistency with the related information. If the validity of the input information cannot be evaluated with the recorded related information alone, a notification to that effect will be given. By teaching whether or not the input information is appropriate by a human, the evaluation based on the related information can be enhanced.
22 to 26 show an example in which input information is evaluated using a knowledge system.
FIG. 22 shows the in-sentence pattern of the recording module that analyzes the sentence concerning the rule and records it as a knowledge system. FIG. 23 analyzes the input information and shows it as an in-sentence pattern. FIG. 24 irradiates the knowledge system with the input information pattern [PI], generates a difference pattern [Δ] of the retrieved related information (rule) pattern [PR], and stores this difference in the information evaluation system The operation of generating an evaluation result by irradiating the recording area is shown. In this example, there is a difference in meaning between the input information and the rule, and a recording module indicating rule violation is excited as an evaluation result.

FIG. 25 to FIG. 26 show an example in which the content of a sentence is evaluated from the meaning of a word included in the inputted sentence. The example of FIG. 25 shows an example in which the purpose is good in common sense but the implemented content is evaluated as bad. The input information is analyzed, and who, what, when, where, why, how, and what was analyzed. Each analyzed sentence element is irradiated to an information evaluation system and evaluated. In this example, the word corresponding to the purpose includes a word that is evaluated as “good”, and the word corresponding to the predicate includes a word that is evaluated as “bad”. Therefore, the evaluation of the input information is “good for the purpose but bad for the implemented content”.
When information is constructed as a knowledge system in this way, problem solving methods can be learned and knowledge can be improved. First, analyze the problem, clarify the problem, and set a goal for problem solving. 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.
Problem analysis and problem clarification are implemented by learning how to detect and identify differences between the target state and the current situation. Moreover, the solution proposal corresponding to each problem is implemented by learning the corresponding processing.

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 generally be expressed as conditional processing. In the present invention, the meaning of the language is interpreted and autonomously converted as conditional processing. The condition of the conditional processing is determined by generating a search pattern from the corresponding word and searching 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 the corresponding information is acquired, the process according to the content is performed.
In the learning phase, when a pattern is excited, the history of the excited pattern is referred to during the set period before the pattern is excited, and the connection relationship with the excited pattern is strengthened. In the practical phase, the history of the excitation pattern in the set period is collated with the record of the connection relationship with other patterns recorded for each recording module, and the pattern is excited when the correlation is large. When the pattern is excited, the history of the excitation pattern is updated, and patterns having a large correlation with the history are sequentially excited in the new state.
In this way, in the learning phase, a series of patterns corresponding to processing is sequentially excited by a human instruction, and by recording the history, the corresponding pattern is excited according to the instructed procedure. The behavior of the pattern is not static but shows dynamic behavior. Using the internal pattern recorded in the pattern, it is possible to search for necessary information and store the search result in a required place. The search result and the condition are collated, and processing according to the condition is performed. Also, it is possible to perform processing such as converting the internal pattern recorded in the pattern into a designated arrangement. This pattern layout conversion is used for language conversion (translation) and utilization of specialized knowledge (formula conversion, chemical formula conversion, physical formula conversion, official conversion). Numerous processes can be implemented by combining patterns that can behave in such a dynamic manner.

FIG. 27 shows an operation for directly executing processing from the input language. In the input sentence (language pattern), the type of sentence is detected, and a processing sequence corresponding to the type of sentence is searched. The searched processing sequence (a series of pattern groups) is stored in the pattern controller, and processing corresponding to the processing sequence is sequentially performed.
FIG. 28 shows an operation of analyzing an input language (question) as an example of a processing sequence and generating an answer to the question autonomously. When it is detected that the input sentence is a question, a search pattern corresponding to the question is generated from the in-sentence pattern included in the input sentence. Since the input sentence has already been analyzed for the subject, predicate, subject modification, predicate modification (what, when, where, why, how), the search pattern asks what information in the sentence element Can be identified. A search pattern is generated from the identification result. Next, the search pattern is irradiated to the pattern recorder, and related information is searched. Of the patterns that show correlation by irradiation of the search pattern, the pattern that shows the highest correlation with the search pattern and has information at the location of the question is stored as the search result. If necessary, it is possible to add processing for confirming whether the search result satisfies the expected answer condition. A pattern corresponding to the answer to the question is generated and output from the search result and the search pattern. The pattern is converted into information (language) by the pattern inverse converter and output.

FIG. 29 shows an operation in which a conditional process is identified from a language input as a process sequence and the process is executed while autonomously confirming the establishment of the condition. By analyzing the input sentence, an in-sentence pattern corresponding to the condition is extracted from the pattern corresponding to the input sentence, and a search pattern [PQ] for the condition search is generated. Search pattern [PQ] in pattern recorder
Search and store search results. It is assumed that information about conditions is recorded in the pattern recorder. If no information on the condition is recorded in the pattern recorder, the fact is notified and an additional input of information is requested. When information corresponding to the condition is retrieved, the information is collated with the condition described in the input sentence. Processing is determined depending on whether or not the collation results match. If the collation matches, the pattern corresponding to the process described in the input sentence is excited and the process is executed.

FIG. 30 shows an operation for advancing the processing while confirming the conditions sequentially for the input series of languages (language group). When processing is in progress, there are ambiguities or ambiguities related to conditional search or execution of processing, and if processing is difficult, notification is made to that effect, and when necessary information is entered, processing is performed according to that information. Let it continue.
In this way, teaching to the autonomous problem solving machine can be performed by sequentially inputting language information corresponding to the processing without sequentially programming each processing. The input language information is analyzed for syntax, meaning, and relationship with already recorded information, and the corresponding pattern is excited according to the analysis result, and processing is executed. As specific processing, various processing such as value evaluation and recording of input information, execution of an instructed command, generation of a solution to a problem / issue can be performed.
The overall operation of this autonomous problem solver 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.).

FIG. 31 shows an operation for detecting a problem and generating a solution autonomously. The state that is the subject of the problem is expressed as a pattern [Xi]. A pattern [X] that is in an X state is extracted from the pattern [Xi]. The state [X] of X is compared with the pattern [XT] indicating the target state, and a difference [ΔX] is detected. The pattern [ΔX] indicating the difference is irradiated to the recording area where the cause / problem analysis pattern is recorded, and the cause / problem causing the difference is extracted. Furthermore, the cause / problem pattern excites the corresponding countermeasure pattern. The countermeasure pattern is stored in the pattern controller, and the countermeasure execution pattern is generated. By irradiating the execution pattern of this countermeasure, the countermeasure is executed. By executing the countermeasure, the state [Xi] is updated. Therefore, the updated state is compared with the target state, and if there is a difference, the process is continued until the difference is eliminated. If indefinite or unclear points occur during the process, such as cause / issue analysis or countermeasures, and if it is difficult to continue the process, notify that and follow the additional information or response instructions given by humans. , Proceed with the process.
As described above, for problems that humans can solve by thinking (language), the autonomous problem-solving machine of the present invention indicates the input problem by indicating the method or procedure for solving the problem in language. The problem can be resolved autonomously according to the solution technique or procedure. 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.

Effect of the invention

According to the present invention, the analysis of the meaning of the input information and the analysis of the relationship with the information recorded in the past are carried out by human instruction or learning, and useful information is associated with each other to construct a knowledge system. It is possible to generate an answer autonomously using a solution taught in the past for the given problem.
Conventionally, when a machine performs processing of input information or the like, it is 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 install 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, 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 addition, regarding problems that humans can solve by thinking (language), the autonomous problem-solving machine of the present invention indicates the method or procedure of input problem solving by indicating the technique or procedure for solving the problem in language. Can solve the problem autonomously.
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.

Configuration example of autonomous problem solver Pattern configuration example Example of operation in which words, sentence elements, and modification relationships are identified sequentially from a word string Example of current affairs Analysis of current affairs (part 1) Analysis of current affairs (part 2) Analysis of current affairs (part 3) Analysis of current affairs (part 4) Analysis of current affairs (part 5) Example questions for current affairs Example of answer generation for questions (configuration diagram) Answer generation for questions (part 1) Answer generation for questions (part 2) Answer generation for questions (part 3) Answer generation for questions (part 4) Answer generation for questions (5) Example of utilization of knowledge system (part 1) (configuration diagram) Example of utilization of knowledge system (part 2) (configuration diagram) Knowledge system utilization example (part 3) (configuration diagram) Example of utilization of knowledge system (part 4) (configuration diagram) Example of how to detect semantic differences Input information evaluation example (1) Input information evaluation example (2) Input information evaluation example (3) (configuration diagram) Input information evaluation example (4) Input information evaluation example (5) (configuration diagram) Example of how to execute processing directly from the input language Example of how to execute the process directly from the input language (answer to the question) Example of how to execute processing directly from the input language (conditional processing) Example of how to execute processing directly from the input language (Sequential execution of language group) Example of how to detect problems and solve them autonomously

1 Pattern Converter 2 Pattern Recorder 3 Pattern Register 4 Pattern Controller 5 Pattern Inverse Converter 6 Pattern Analyzer 7 Pattern Irradiator

Claims (5)

A pattern converter that converts information into a pattern, a pattern recorder composed of a pattern, a connection between patterns, and a recording module that records the relationship between patterns, and a pattern or a connection between patterns. A pattern register that automatically registers and changes, a pattern controller that controls pattern processing, and records the pattern excitation history, and when the pattern is excited, the pattern from the pattern history data previously excited The connection relationship data is generated and recorded in the connection relationship recording unit of the recording module of the pattern, and the excitation pattern history data set from the current time to the past and the connection recorded in the connection relationship recording unit of each recording module A pattern irradiator that collates the relevant data and excites a recording module with a high correlation; A pattern inverse converter that converts patterns into information and a pattern analyzer that analyzes patterns and relationships between patterns. Subject, subject modifiers, predicates, predicate modifiers, modifier relationships, when and where Artificial intelligence and software that analyzes and records who, what, how and why. A pattern converter that converts information into a pattern, a pattern recorder composed of a pattern, a connection between patterns, and a recording module that records the relationship between patterns, and a pattern or a connection between patterns. A pattern register that automatically registers and changes, a pattern controller that controls pattern processing, and records the pattern excitation history, and when the pattern is excited, the pattern from the pattern history data previously excited The connection relationship data is generated and recorded in the connection relationship recording unit of the recording module of the pattern, and the excitation pattern history data set from the current time to the past and the connection recorded in the connection relationship recording unit of each recording module A pattern irradiator that collates the relevant data and excites a recording module with a high correlation; A pattern inverse converter that converts patterns into information and a pattern analyzer that analyzes patterns and relationships between patterns. Subject, subject modifiers, predicates, predicate modifiers, modifier relationships, when and where Analyze who, what, how, why, and artificially generate answers to questions using truth, truth, facts, expertise, rules, and common sense data built into pattern recorders Intelligence and software. A pattern converter that converts information into a pattern, a pattern recorder composed of a pattern, a connection between patterns, and a recording module that records the relationship between patterns, and a pattern or a connection between patterns. A pattern register that automatically registers and changes, a pattern controller that controls pattern processing, and records the pattern excitation history, and when the pattern is excited, the pattern from the pattern history data previously excited The connection relationship data is generated and recorded in the connection relationship recording unit of the recording module of the pattern, and the excitation pattern history data set from the current time to the past and the connection recorded in the connection relationship recording unit of each recording module A pattern irradiator that collates the relevant data and excites a recording module with a high correlation; A pattern inverse converter that converts patterns into information and a pattern analyzer that analyzes patterns and relationships between patterns. Subject, subject modifiers, predicates, predicate modifiers, modifier relationships, when and where Analyzing who, what, how and why, and using the truth, truth, facts, expertise, rules and common sense data built on the pattern recorder to evaluate information, Artificial intelligence and software to record as a knowledge system. A pattern converter that converts information into a pattern, a pattern recorder composed of a pattern, a connection between patterns, and a recording module that records the relationship between patterns, and a pattern or a connection between patterns. A pattern register that automatically registers and changes, a pattern controller that controls pattern processing, and records the pattern excitation history, and when the pattern is excited, the pattern from the pattern history data previously excited The connection relationship data is generated and recorded in the connection relationship recording unit of the recording module of the pattern, and the excitation pattern history data set from the current time to the past and the connection recorded in the connection relationship recording unit of each recording module A pattern irradiator that collates the relevant data and excites a recording module with a high correlation; A pattern inverse converter that converts patterns into information and a pattern analyzer that analyzes patterns and relationships between patterns. Subject, subject modifiers, predicates, predicate modifiers, modifier relationships, when and where Analyzing who, what, how and why, and using the truth, truth, facts, expertise, rules, common sense data and information obtained in the pattern recorder in the language Artificial intelligence and software that autonomously performs the indicated process. A pattern converter that converts information into a pattern, a pattern recorder composed of a pattern, a connection between patterns, and a recording module that records the relationship between patterns, and a pattern or a connection between patterns. A pattern register that automatically registers and changes, a pattern controller that controls pattern processing, and records the pattern excitation history, and when the pattern is excited, the pattern from the pattern history data previously excited The connection relationship data is generated and recorded in the connection relationship recording unit of the recording module of the pattern, and the excitation pattern history data set from the current time to the past and the connection recorded in the connection relationship recording unit of each recording module A pattern irradiator that collates the relevant data and excites a recording module with a high correlation; A pattern inverse converter that converts patterns into information and a pattern analyzer that analyzes patterns and relationships between patterns. Subject, subject modifiers, predicates, predicate modifiers, modifier relationships, when and where Analyzing who, what, how, why, and the truth, truth, facts, expertise, rules, common sense data, information obtained, and language built on the pattern recorder for the specified problem Artificial intelligence and software that solves problems autonomously using the countermeasures taught.