JPH05342249A - Statistic analysis controller - Google Patents

Statistic analysis controller

Info

Publication number
JPH05342249A
JPH05342249A JP15166192A JP15166192A JPH05342249A JP H05342249 A JPH05342249 A JP H05342249A JP 15166192 A JP15166192 A JP 15166192A JP 15166192 A JP15166192 A JP 15166192A JP H05342249 A JPH05342249 A JP H05342249A
Authority
JP
Japan
Prior art keywords
data
analysis
result
analyzed result
unit
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.)
Withdrawn
Application number
JP15166192A
Other languages
Japanese (ja)
Inventor
Toru Kishimoto
徹 岸本
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.)
KOBE NIPPON DENKI SOFTWARE KK
NEC Software Kobe Ltd
Original Assignee
KOBE NIPPON DENKI SOFTWARE KK
NEC Software Kobe Ltd
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 KOBE NIPPON DENKI SOFTWARE KK, NEC Software Kobe Ltd filed Critical KOBE NIPPON DENKI SOFTWARE KK
Priority to JP15166192A priority Critical patent/JPH05342249A/en
Publication of JPH05342249A publication Critical patent/JPH05342249A/en
Withdrawn legal-status Critical Current

Links

Abstract

PURPOSE:To recognize an analyzed result by Japanese by displaying the analyzed result by means of a statistic analysis part and a judged result by means of an analyzed result judgement part by means of combining them. CONSTITUTION:Input data 1 is data of a measured table system, and it is the set of data on a numeric value, a character and a code. A data input part 2 fetches a part required for analysis from input data 1, and transfers control to the statistic analysis part 3. The statistic analysis part 3 analyzes data based on a designated statistic method, and the analyzed result is outputted to which an identification index is added. The analyzed result judgement part 5 verifies whether the analyzed result is statistically correct or not based on information accumulated in a knowledge base 6, executes synthetic judgement and outputs judged result data 7. An editing/display part 8 combines analyzed result data 4 with judged result data 7, edits a layout so that a man can easy to view and displays display data 9.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【産業上の利用分野】本発明は、統計解析システムに関
し、特に、統計解析制御装置に関する。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a statistical analysis system, and more particularly to a statistical analysis control device.

【0002】[0002]

【従来の技術】従来、統計解析制御装置は、図4に示す
ように、解析した結果をそのまま数値として出力した
り、一部グラフ化して表示しているが、判断はすべて人
間に任されている。
2. Description of the Related Art Conventionally, a statistical analysis control device, as shown in FIG. 4, outputs an analysis result as it is as a numerical value or displays it as a part of a graph, but it is entirely up to humans to make decisions. There is.

【0003】[0003]

【発明が解決しようとする課題】上述した従来の統計解
析制御装置では、表示がわかりにくい、統計の専門的知
識が必要であるという欠点がある。
The above-described conventional statistical analysis control device has the drawbacks that the display is difficult to understand and specialized statistical knowledge is required.

【0004】[0004]

【課題を解決するための手段】本発明の統計解析制御装
置は、データを入力するデータ入力部と、指定された統
計手法に基づきデータ解析を行なう統計解析部と、前記
統計解析部による解析結果を知識ベースに蓄積されてい
る情報に基づき統計的に検証し総合的な判断を下す解析
結果判断部と、前記統計解析部による解析結果および前
記解析結果判断部による判断結果を組合せて表示する編
集・表示部とを具備することを特徴とする。
A statistical analysis control device of the present invention comprises a data input section for inputting data, a statistical analysis section for performing data analysis based on a specified statistical method, and an analysis result by the statistical analysis section. Edit to display a combination of the analysis result judgment unit that statistically verifies the information based on the information accumulated in the knowledge base and makes a comprehensive judgment, and the analysis result by the statistical analysis unit and the judgment result by the analysis result judgment unit -A display unit is provided.

【0005】[0005]

【実施例】次に、本発明について図面を参照して説明す
る。
DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described with reference to the drawings.

【0006】図1は本発明の一実施例のブロック図であ
る。入力データ1は、測定された表形式のデータであ
り、数値,文字,コードなどのデータの集合となってい
る。データ入力部2は、入力データ1から分析に必要な
部分を取り込み、統計解析部3に制御を渡す。統計解析
部3は、従来と同様に指定された統計手法に基づき、デ
ータの解析を行なう。解析結果は、識別用インデックス
を付加されて、出力される。解析結果判断部5は、知識
ベース6に蓄積されている情報に基づき、解析された結
果が統計的に正しいか否かを検証し、その結果に基づい
て、さらに総合的な判断を下し、判断結果データ7を出
力する。編集・表示部8は、解析結果データ4と判断結
果データ7とを組み合わせて、人間が見やすいようにレ
イアウトを編集し、最終的な表示データ9を表示する。
FIG. 1 is a block diagram of an embodiment of the present invention. The input data 1 is the measured tabular data, which is a set of data such as numerical values, characters, and codes. The data input unit 2 takes in a portion required for analysis from the input data 1 and passes control to the statistical analysis unit 3. The statistical analysis unit 3 analyzes the data based on the designated statistical method as in the conventional case. The analysis result is output with the identification index added. The analysis result judging unit 5 verifies whether or not the analyzed result is statistically correct based on the information accumulated in the knowledge base 6, and makes a more comprehensive judgment based on the result. The judgment result data 7 is output. The editing / display unit 8 combines the analysis result data 4 and the judgment result data 7 to edit the layout so that it is easy for humans to see, and displays the final display data 9.

【0007】図2は、知識ベース6に蓄積されている情
報に基づき、チェックと判定を行なう解析結果判断部5
を示すブロック図である。統計解析部3で解析され、出
力された解析結果データ4には、識別用インデックスが
付与されている。この識別用インデックスに従い、ま
ず、原データの持つ特性(例えば、母集団の大きさ、デ
ータの片寄り、異常値の有無、データのクリーニングな
ど)の妥当性の検証をデータの妥当性判定部51で行な
う。次に変数間に相関関係の有無の検証を変数の独立性
判定部52で行なう。さらに、回帰モデルなどのモデル
式の妥当性の検証をモデルの妥当性判定部53で行な
う。最後に各判定部51〜53の検証結果に加えて、人
間の経験則を加味されている知識ベース6の情報をも加
味して、総合的な判断を総合判断部54で行なう。
FIG. 2 shows an analysis result judging section 5 for checking and judging based on the information accumulated in the knowledge base 6.
It is a block diagram showing. The analysis result data 4 analyzed and output by the statistical analysis unit 3 is provided with an identification index. According to the identification index, first, the validity of the characteristics of the original data (for example, the size of the population, the deviation of the data, the presence / absence of an abnormal value, the cleaning of the data, etc.) is verified, and the data validity determining unit 51. Do in. Next, the variable independence determination unit 52 verifies the presence or absence of correlation between variables. Further, the model validity determining unit 53 verifies the validity of the model formula such as the regression model. Finally, in addition to the verification results of the determination units 51 to 53, the comprehensive determination unit 54 makes a comprehensive determination by also considering the information of the knowledge base 6 in which human empirical rules are added.

【0008】具体的なデータに基づいて実行例を次に示
す。
An example of execution is shown below based on specific data.

【0009】データ入力部2を通して入力されたデータ
が以下のようなものとし、 生徒番号 数学 理科 国語 英語 1 70 85 75 70 2 90 95 60 80 3 40 50 90 70 : : : : : : : : : : “数学”の成績と“理科”の成績(得点)との関連性に
ついて分析する場合、まず、統計解析部3で単回帰分析
を行ない、識別用インデックスを付与した標準統計量や
回帰係数、分散分析表などを作成する。次に、知識ベー
ス6を参照しながら、データの分布と標準偏差からデー
タの片寄りと異常値をチェックし、データな妥当性をチ
ェックする(データの妥当性判定部51)。さらに相関
係数とそのt検定結果より変数の独立性をチェック(変
数の独立性判定部52)し、分散分析表の変動因子の値
とF検定結果よりモデル式の妥当性をチェックする(モ
デルの妥当性判定部53)。
It is assumed that the data input through the data input unit 2 is as follows: Student number Mathematics Science Japanese English 1 70 85 75 75 70 2 90 95 60 60 80 3 40 50 90 70: :: :: :: :: :: : When analyzing the relationship between the results of "mathematics" and the results (scores) of "science", first, a simple regression analysis is performed in the statistical analysis unit 3, and standard statistics or regression coefficients with identification indexes are given. Create an ANOVA table, etc. Next, referring to the knowledge base 6, the deviation of the data and the abnormal value are checked from the data distribution and the standard deviation, and the validity of the data is checked (the validity judgment unit 51 of the data). Further, the independence of variables is checked from the correlation coefficient and its t-test result (variable independence determination unit 52), and the validity of the model formula is checked from the values of the variable factors in the analysis of variance table and the F-test results (model Validity determination section 53).

【0010】最後に各判定部51〜53での検証結果に
知識ベース6に蓄積された人間の経験則(例:このよう
なデータの場合、寄与率が0.8%以上あることが望ま
しいなど)を加味して、総合的な判断結果を下す(総合
判断部54)。
Finally, human verification rules accumulated in the knowledge base 6 in the verification results of the determination units 51 to 53 (eg, in the case of such data, it is desirable that the contribution rate is 0.8% or more). ) Is added to give a comprehensive judgment result (general judgment unit 54).

【0011】総合的な判断結果としては、次の4種類に
分け、 有意水準 評価メッセージ 1%以下 かなり強い関連がある 1〜5% かなり関連がある 5〜10% やや関連がある 10%以上 ほとんど関連がない それぞれに対応する日本語メッセージを編集・表示部8
で解析データとも組み合わせて分かりやすく表示する。
The overall judgment result is divided into the following four types: Significance level evaluation message 1% or less Significantly related 1-5% Significantly related 5-10% Somewhat related 10% or more Most Unrelated Japanese messages corresponding to each are edited / displayed 8
It is also combined with analysis data and displayed in an easy-to-understand manner.

【0012】図3は、半回帰分析の解析結果と解析結果
判断部5で総合的に判断された結果を編集・表示部8で
人間にわかりやすい形(グラフ、日本語)に直して表示
した例を示す。
FIG. 3 shows an example in which the analysis result of the semi-regression analysis and the result comprehensively judged by the analysis result judging unit 5 are displayed on the editing / display unit 8 in a form (graph, Japanese) which is easy for humans to understand. Indicates.

【0013】この評価メッセージやグラフにより、一見
して数学と理科の成績の関連性をつかむことができる。
With this evaluation message and graph, it is possible to grasp at first glance the relationship between mathematics and the results of science.

【0014】[0014]

【発明の効果】以上説明したように、本発明は、統計解
析結果に対して制御装置側で必要な統計的判断を総合的
に行なうことで解析結果が日本語で一目で分かり、統計
の知識がなくても使えるという効果を奏する。
As described above, according to the present invention, the analysis result can be understood at a glance in Japanese by statistically making the necessary statistical judgment on the control device side with respect to the statistical analysis result. The effect is that it can be used even without.

【図面の簡単な説明】[Brief description of drawings]

【図1】本発明の一実施例のブロック図である。FIG. 1 is a block diagram of an embodiment of the present invention.

【図2】図1中の解析結果判断部のブロック図である。FIG. 2 is a block diagram of an analysis result determination unit in FIG.

【図3】図1の実施例での総合的判断結果による具体的
表示例を示す図である。
FIG. 3 is a diagram showing a specific display example based on a comprehensive judgment result in the embodiment of FIG.

【図4】従来例による出力結果を示す図である。FIG. 4 is a diagram showing an output result according to a conventional example.

【符号の説明】[Explanation of symbols]

1 入力データ 2 データ入力部 3 統計解析部 4 解析結果データ 5 解析結果判断部 6 知識ベース 7 判断結果データ 8 編集・表示部 9 表示データ 1 Input data 2 Data input part 3 Statistical analysis part 4 Analysis result data 5 Analysis result judgment part 6 Knowledge base 7 Judgment result data 8 Editing / display part 9 Display data

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】 データを入力するデータ入力部と、指定
された統計手法に基づきデータ解析を行なう統計解析部
と、前記統計解析部による解析結果を知識ベースに蓄積
されている情報に基づき統計的に検証し総合的な判断を
下す解析結果判断部と、前記統計解析部による解析結果
および前記解析結果判断部による判断結果を組合せて表
示する編集・表示部とを具備することを特徴とする統計
解析制御装置。
1. A data input section for inputting data, a statistical analysis section for performing data analysis based on a specified statistical method, and a statistical analysis based on information accumulated in a knowledge base for analysis results by the statistical analysis section. The statistical analysis is characterized by comprising: an analysis result judging unit that verifies and makes a comprehensive judgment; and an editing / display unit that displays the analysis result by the statistical analysis unit and the judgment result by the analysis result judging unit in combination. Analysis control device.
JP15166192A 1992-06-11 1992-06-11 Statistic analysis controller Withdrawn JPH05342249A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP15166192A JPH05342249A (en) 1992-06-11 1992-06-11 Statistic analysis controller

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP15166192A JPH05342249A (en) 1992-06-11 1992-06-11 Statistic analysis controller

Publications (1)

Publication Number Publication Date
JPH05342249A true JPH05342249A (en) 1993-12-24

Family

ID=15523462

Family Applications (1)

Application Number Title Priority Date Filing Date
JP15166192A Withdrawn JPH05342249A (en) 1992-06-11 1992-06-11 Statistic analysis controller

Country Status (1)

Country Link
JP (1) JPH05342249A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002202115A (en) * 2000-11-09 2002-07-19 Samsung Electronics Co Ltd Method of automatically detecting measuring error of measuring apparatus

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002202115A (en) * 2000-11-09 2002-07-19 Samsung Electronics Co Ltd Method of automatically detecting measuring error of measuring apparatus

Similar Documents

Publication Publication Date Title
JPH05342249A (en) Statistic analysis controller
SE9801964D0 (en) Simulation system
JP2010217862A (en) Electronic learning monitoring method and system
JP2002351522A (en) System and method for supporting reduction of line balance loss, and recording medium stored with program for realizing the method
JP2002254263A (en) System for evaluating workability of assembly
JPH11353301A (en) Method for evaluating human error analysis
JP6878244B2 (en) Driving operation evaluation system
CN112599215B (en) Rule adjustment method, device, equipment and storage medium of automatic auditing system
KR920003302B1 (en) Buzzer circuit for cash register
JPH0519843A (en) Digital controller
JP3211235B2 (en) Approximate reasoning device and human interface device
JPH08328644A (en) Method for analyzing fluctuation in process
JPH07262056A (en) Monitor device for failure occurrence trend of computer
JPH04264921A (en) Project evaluator for software development
CN113127757A (en) User data recommendation method and device, readable storage medium and electronic equipment
JP2005309936A (en) Operation load measurement device
JPH05144941A (en) Data treatment device
JP2001257146A (en) Measuring apparatus, processing, apparatus and method for measurement and processing
JP2000194585A (en) Execution time calculating device for program
GB2139767A (en) Fault diagnosis in an industrial operation
JPH06290080A (en) Remote maintenance device
JPH07262014A (en) Diagnostic type expert system
JPS63131239A (en) Method for deciding quality of program
JPS6220030A (en) High-speed program shaping output device
JPS62293354A (en) Back-up device formation of specialist system

Legal Events

Date Code Title Description
A300 Withdrawal of application because of no request for examination

Free format text: JAPANESE INTERMEDIATE CODE: A300

Effective date: 19990831