JP7211600B1 - Information analysis system - Google Patents

Abstract

Kind Code: A1 A novel information analysis system is provided that improves convenience for a user who extracts and analyzes accident information. Kind Code: A1 An information analysis system used when analyzing an accident using an accident information database in which accident information characterized by a plurality of attribute items is accumulated, wherein a first group presenting means for presenting a first accident information group narrowed down by a specified attribute item; an attribute item presenting means for presenting an attribute item for further narrowing down said first accident information group; a second group presenting means for presenting a second accident information group narrowed down by the attribute items selected by the attribute item presenting means. [Selection drawing] Fig. 1

Description

The present invention relates to an information analysis system.

A large number of failure/accident information (hereinafter simply referred to as "accident information") has been reported, such as accidents caused by failures and defects of electrical and mechanical products, etc., and accidents caused by usage. are stored in databases held by companies and government agencies.

Patent Literature 1 describes a technique for extracting desired accident information using text mining in order to analyze the cause and details of an accident using the accident information accumulated in such a database.

JP 2020-194531 A

An object of the present invention is to provide an unprecedented information analysis system that improves convenience for users who extract and analyze accident information.

The gist of the present invention will be explained.

An information analysis system used when analyzing an accident using an accident information database in which accident information characterized by a plurality of attribute items is accumulated,
first group presenting means for presenting a first accident information group narrowed down by specified attribute items from arbitrarily selected accident information groups;
attribute item presentation means for presenting attribute items for further narrowing down the first accident information group;
a second group presenting means for presenting a second accident information group narrowed down by the attribute items selected by the attribute item presenting means ;
The means for presenting each group counts the number of accidents belonging to each accident information group, calculates the frequency or probability of occurrence of an accident as an accident information group, and uses the frequency or probability of occurrence as a common axis. An information analysis system characterized in that it is configured to display information groups .

Also, an information analysis system used when analyzing an accident using an accident information database in which accident information characterized by a plurality of attribute items is accumulated,
first group presenting means for presenting a first accident information group narrowed down by specified attribute items from arbitrarily selected accident information groups;
attribute item presentation means for presenting attribute items for further narrowing down the first accident information group;
a second group presenting means for presenting a second accident information group narrowed down by the attribute items selected by the attribute item presenting means;
and risk matrix display means for arranging the accident information groups in corresponding cells of a risk matrix and displaying the number of accident information arranged in each cell. .

Further, in the information analysis system according to claim 1 , the occurrence frequency or occurrence probability is automatically calculated from a denominator obtained according to the attribute item specified when narrowing down the first accident information group. The present invention relates to an information analysis system characterized by having automatic calculation means.

Further, in the information analysis system according to any one of claims 1 to 3, the attribute item presentation means is configured to display a list of attribute items that enable further narrowing down of the presented accident information group. The present invention relates to an information analysis system characterized by:

Since the present invention is configured as described above, it becomes an unprecedented information analysis system that improves convenience for users who extract and analyze accident information.

It is a flow chart of a present Example. It is a schematic explanatory drawing of the hazard matrix of a present Example. FIG. 10 is a schematic explanatory diagram of narrowing down analysis of the present embodiment; FIG. 10 is a schematic explanatory diagram of narrowing down analysis of the present embodiment; FIG. 10 is a schematic explanatory diagram of narrowing down analysis of the present embodiment; FIG. 10 is a schematic explanatory diagram of narrowing down analysis of the present embodiment; FIG. 10 is a schematic explanatory diagram of narrowing down analysis of the present embodiment; It is a schematic explanatory drawing of the search result display means of a present Example. It is a schematic explanatory drawing of a risk matrix.

A preferred embodiment of the present invention will be briefly described with reference to the drawings showing the operation of the present invention.

When retrieving and analyzing accident information from the accumulated accident database, the first accident information group is displayed by narrowing down the specified attribute items from the accident information groups presented on the screen. be.

At this time, when an attribute item presented by the attribute item presenting means is selected, a second accident information group obtained by further narrowing down the first accident information group is presented. By repeating this process, it becomes possible to interactively identify specific aspects of accident risk through dialogue with the system. That is, the user can clarify the specific contents of the accident by repeatedly executing the attribute item presentation means for narrowing down and the narrowed down second group presentation means following the first group presentation means.

A specific embodiment of the present invention will be described with reference to the drawings.

This embodiment is an information analysis system used when analyzing an accident using an accident information database in which accident information characterized by a plurality of attribute items is accumulated.

Specifically, a first group presentation means for presenting a first accident information group narrowed down by a specified attribute item from arbitrarily selected accident information groups; An attribute item presenting means for presenting attribute items for narrowing down, and a second group presenting means for presenting a second accident information group narrowed down by the attribute items selected by the attribute item presenting means. .

Each group presentation means is configured to display the accident information groups side by side on the axis based on the values (divisions) of the accident occurrence frequency or occurrence probability as a common axis.

Further, it has automatic calculation means for automatically calculating the frequency of occurrence or the probability of occurrence from a denominator obtained according to the attribute item specified when narrowing down the first accident information group. For example, if an attribute item such as population, gender, or region is selected, the population parameter with that attribute will be the breakdown of the original accident group, so it is necessary to recalculate the frequency or probability of occurrence. This recalculation is done automatically.

Further, the attribute item presentation means is configured to display a list of all attribute items that enable further narrowing down of the presented accident information group.

Further, this embodiment includes tag information presenting means for presenting tag information attached to accident information together with the number of accident information related to the tag information, and tag information related to tag information selected from the presented tag information. and search result display means for displaying search results of extracted accident information.

This embodiment includes a control device such as a CPU, a main storage device (memory) such as a RAM, an auxiliary storage device such as an HDD or SSD, a display device such as a liquid crystal display device, and an input device such as a keyboard or mouse. , a communication device such as a NIC for connecting to a communication network, and a general computer provided with a bus. For example, the functions of this embodiment are stored in the form of a program in an auxiliary storage device of a computer. The program is read from the auxiliary storage device, expanded on the memory, and executed by the CPU. This system may be composed of a single computer, or may be composed of a server and at least one computer connected to this server via a network (part of the functions are may run on a connected server).

Schematically, in this embodiment, when searching and extracting accident information that includes a selected attribute item (tag information) by referring to an accident information database, and sequentially narrowing down with different tag information, the search The tag information included in a predetermined group of the accident information obtained is displayed side by side.

FIG. 1 shows a flow chart of this embodiment. In the flowchart, the analysis can be performed by performing at least "Import matter" -> "Edit matter" -> "Creating hazard matrix" -> "Analysis narrowing down" in this order.

Auxiliary storage device stores product accident information, recall information, customer feedback (requests and complaints) and repair information from the Ministry of Economy, Trade and Industry and the Consumer Affairs Agency in advance, as well as customer feedback and complaint information held by companies. Create an accident information database (item import). In addition, definition and editing of vocabulary used for tag information and the like may be performed in advance (vocabulary editing).

Vocabulary used for tag information can be based on IIDF (Injury Information Description Framework Coding Manual) or AIDF (Accident Information Description Framework Coding Manual). The IIDF is a framework consisting of five elements, which are the four elements of the epidemiological model proposed by WHO, plus failure results.

The acquired information is appropriately associated with the tag information by the item editing means (item editing). Also, if necessary, tag editing is performed by the tag information editing means.

For example, when editing a case, it is not limited to manual editing. Predetermined words are automatically tagged based on the words written in the accident description column and recurrence prevention countermeasure column in the accident information (case). It is good also as a structure which associates with tag information automatically. Also, when a predetermined wording is associated with predetermined tag information by tag editing (used as teaching data for AI), if the wording is automatically included thereafter, it is associated with the tag information. Also good.

A hazard matrix is created and displayed by performing aggregation based on the tag information associated with the project edit (creation of a hazard matrix). For example, in a matrix where the vertical axis is the frequency of occurrence and the horizontal axis is the degree of harm, the number of cases is displayed for each combination of hazard and mechanism (see Fig. 2).

In the hazard matrix, when analyzing the accident information contained in the attribute items of one selected column, select the risk map button under each column (see FIGS. 3 to 7). Then, when one attribute item (in this embodiment, a combination of a hazard and a mechanism) is selected in the column, a tag information display area and a pull-down menu are displayed on the right side, and one attribute item is selected from the pull-down menu. Then, the tag information included in the attribute item is displayed in the tag information display area. By repeating in the same way, it is possible to narrow down the cases in the attribute item. This narrowing operation can be performed one or more times (narrowing analysis). In addition, the narrowing down operation is configured so that it is possible to return to any stage and reselect the attribute item. Further, by selecting the presented tag information, a list of accident information including the tag information can be displayed (see FIG. 8).

A detailed description of the refinement analysis will be given. When the risk map button in FIG. 2 is selected, the risk scope screen is displayed. The risk scope screen is a screen in which the degree of harm selected on the vertical axis (common arrangement axis) and horizontal axis of the hazard matrix is displayed on the left side of one column of the hazard matrix.

In this state, when one attribute item (H3.0M1 in FIG. 2) is selected, a tag information display area and a pull-down menu are displayed on the right side, and one attribute item (part of failure in FIG. 2) is displayed from the pull-down menu. is selected, the tag information in the attribute item is displayed together with the number of each item.

Subsequently, when one piece of tag information (wrist and hand in FIG. 2) is selected from the attribute items and the confirmation button is selected, a tag information display area and a pull-down menu are additionally displayed on the right side. If you select the action when an accident occurs as an attribute item from the pull-down menu, the tag information in the attribute item will be displayed in the same way. Furthermore, a tag information display area and a pull-down menu are additionally displayed.

In the same manner, select "Treatment classification", "Hospitalization in the same hospital after treatment", "Disabled alcoholic beverages", "Burns, electricity, etc.", and "Roles in accidents" and perform analysis.

In addition, FIG. 8 shows a state in which "grip" is selected and a list of accident information including "grip" tag information is displayed.

Each means for performing the above processing will be described below.

The attribute item presentation means displays the tag information classified into groups such as the degree of harm of the event, the type of injury, the site of injury, the classification of treatment, the role in the accident, and the action at the time of occurrence of the accident for each group (accident information contains at least one group of tag information). In this embodiment, selection is made from the pull-down menu below the tag information display area by the tag information presentation means.

The tag information presenting means displays, in the tag information display area, the aggregated result for each piece of tag information included in one attribute item selected from the attribute items presented by the attribute item presenting means.

The search result display means selects the displayed tag information and displays a list of accident information including the tag information. In this embodiment, it is displayed below the tag information display area and the pull-down menu.

In this embodiment, the attribute item presenting means can present the attribute items at a plurality of locations, and the tag information presenting means displays the tag information included in each of the groups selected at different locations side by side. It is configured to allow

Specifically, attribute item presentation start means (another pull-down menu and tag means for displaying the information display area on the right).

The accident information is classified for each attribute item, and has attribute item presentation means for presenting the attribute items. Extract the accident information related to the information.

In this embodiment, the attribute item includes a combination of the type of "danger" and the type of "mechanism". This assigns a code to H indicating the hazard and M indicating the mechanism, respectively, and is displayed as HxxxMyyy (zzz). For example, H3M5 (12) displayed in the hazard matrix means that there are 12 accident information items with attribute items of 3 hazards and 5 mechanisms.

The present embodiment also has analysis target selection means for selecting in advance accident information to be analyzed from the accident information in the accident information database, and the tag information presentation means is selected by the analysis target selection means. It is configured to present the tag information selected in the accident information.

Specifically, it has a hazard matrix creation means for creating a hazard matrix by selecting accident information in an arbitrary range from the accident information database, and the analysis target selection means selects any column in the hazard matrix. is configured so that the accident information included in the can be selected as an analysis target. In this embodiment, as described above, the number of cases is displayed in a matrix in which the vertical axis is the frequency of occurrence and the horizontal axis is the degree of harm, and the total number of cases is displayed for each combination (attribute item) of the hazard and the mechanism.

Also, this embodiment is configured to be able to create and display a risk matrix.

For example, using the degree of harm set for the case, the vertical axis is the frequency of occurrence (occurrence frequency classification for each industry), the horizontal axis is the degree of harm, and the number of accident information is displayed in a matrix. The first accident information may be narrowed down from this state. Further, by selecting the number of cells (square), the details of the target accident information can be displayed as a list (see FIG. 9).

Moreover, it is good also as a structure provided with the following functions.

・Map analysis Aggregate and display the number of accidents by prefecture on a map.

・Graph analysis The number of accidents is aggregated and displayed in bar graphs and pie charts by time of occurrence, area, source of danger, mechanism, action at the time of occurrence, place of occurrence, presence or absence of fire, category of accident cause, age, and gender.

・Event drop analysis Analyze trends in the number of accidents by prefecture from 1980 to the present.

・Heat map analysis The degree of occurrence is displayed in different colors for each degree of danger axis and mechanism, presence or absence of fire, place of occurrence, and type of injury.

Since the present embodiment is configured as described above, it becomes an unprecedented information analysis system that improves the convenience for the user who extracts and analyzes the accident information.

Claims (4)

An information analysis system used when analyzing an accident using an accident information database in which accident information characterized by a plurality of attribute items is accumulated,
first group presenting means for presenting a first accident information group narrowed down by specified attribute items from arbitrarily selected accident information groups;
attribute item presentation means for presenting attribute items for further narrowing down the first accident information group;
a second group presenting means for presenting a second accident information group narrowed down by the attribute items selected by the attribute item presenting means ;
The means for presenting each group counts the number of accidents belonging to each accident information group, calculates the frequency or probability of occurrence of an accident as an accident information group, and uses the frequency or probability of occurrence as a common axis. An information analysis system configured to display information groups . An information analysis system used when analyzing an accident using an accident information database in which accident information characterized by a plurality of attribute items is accumulated,
first group presenting means for presenting a first accident information group narrowed down by specified attribute items from arbitrarily selected accident information groups;
attribute item presentation means for presenting attribute items for further narrowing down the first accident information group;
a second group presenting means for presenting a second accident information group narrowed down by the attribute items selected by the attribute item presenting means;
and risk matrix display means for arranging the accident information groups in corresponding cells of a risk matrix and displaying the number of accident information arranged in each cell. 2. The information analysis system according to claim 1 , wherein said frequency of occurrence or probability of occurrence is automatically calculated from a denominator obtained according to said attribute item specified when said first accident information group is narrowed down. An information analysis system characterized by having means. 4. The information analysis system according to any one of claims 1 to 3, wherein said attribute item presentation means is configured to display a list of attribute items that enable further narrowing down of the presented accident information group. An information analysis system characterized by

Images