JP2022014388A - Analyzer, method for analysis, analysis program, analysis system, and terminal device - Google Patents

Abstract

To efficiently grasp the contents of reports in disaster contingency planning.SOLUTION: A topic checking unit 123 extracts the difference between texts in which reports on disasters which happened at different times on different days are written in a freely selected manner. A positive-negative determination unit 124 determines whether the difference extracted by the topic checking unit 123 shows a deterioration of a situation or an improvement of a situation.SELECTED DRAWING: Figure 1

Description

The present invention relates to a text analysis technique.

In recent years, organizations dealing with disasters due to the frequent occurrence of large-scale disasters have been struggling to understand the damage situation and the progress of recovery activities from the damage. Traditionally, the Commissioner of the Fire Service Agency has used "Form No. 4" when requesting reports from local governments (Reference: Notification of revision of disaster report handling guidelines (https://www.fdma.go). .jp / laws / tutatsu / items / 190425oukyuu28% 281% 29.pdf)).

Form No. 4 is a form for collecting various types of information, but it is rarely reported in a comprehensive manner, and it is doubtful that the collected information is fully utilized for disaster response (references). : Paper on the utilization of Form No. 4 (https://www.jstage.jst.go.jp/article/jscejsp/69/2/69_I_121/_pdf/-char/ja)). Therefore, reporting work may be a burden at the site where the response in front of you is busy.

On the other hand, in disaster response, there are various forms of damage according to the type of disaster and the scale of the disaster, and there is a problem that the reporting format cannot be standardized. Therefore, a system has also been developed to report the contents of reports that cannot be standardized in atypical free-form texts (Reference: Application of crisis response support system to large-scale international sporting events-8th Sapporo Asian Winter Games). Utilization of "KADAN" at the Games-, IEICE Journal D Vol.J101-D No.10 pp.1405-1413). Free-form reporting has the advantages of being able to report flexibly according to the situation and saving the trouble of preparing a standard format at the beginning of system introduction.

In order to analyze free-form sentences, it is conceivable to use techniques such as topic classification and trend analysis in the field of natural language (see, for example, Non-Patent Document 1). Topic classification and trend analysis capture the statistical frequency of appearance of character strings contained in sentences and the change in frequency over time.

Bing Liu, Sentiment Analysis: Mining Opinions, Sentiments, and Emotions Cambridge University Press 2015, ISBN978-1-107-01789-4

However, the conventional technique has a problem that it may be difficult to efficiently grasp the contents of the report in disaster countermeasures. For example, the disaster response headquarters may receive reports of free descriptions about the same subject in chronological order. In such a case, the technique described in Non-Patent Document 1 can obtain statistical information such as the frequency of appearance of words appearing in the report, but how the content of the report changes over time. It is difficult to know what it is. Changes in the content of the report may be caused by deterioration or improvement of the situation, and understanding them leads to efficient situation judgment.

According to the present invention, it is possible to efficiently grasp the contents of a report in disaster countermeasures.

FIG. 1 is a diagram showing a configuration example of an analysis system according to the first embodiment. FIG. 2 is a diagram showing an example of a business plan screen. FIG. 3 is a diagram showing an example of a task management screen. FIG. 4 is a diagram showing an example of a text in which a report is written. FIG. 5 is a diagram showing an example of a topic division method. FIG. 6 is a diagram showing an example of a method of combining sentences. FIG. 7 is a diagram showing an example of a positive / negative analysis method. FIG. 8 is a diagram showing an example of a positive / negative analysis method. FIG. 9 is a diagram showing an example of a risk detection table. FIG. 10 is a diagram that gives a bird's-eye view of changes in each topic. FIG. 11 is a diagram showing an example of a graph showing the result of comparing texts. FIG. 12 is a diagram showing an example of the result of comparing texts. FIG. 13 is a diagram showing an example of the result of comparing texts. FIG. 14 is a diagram showing an example of the result of comparing texts. FIG. 15 is a diagram showing a processing flow of the analysis system according to the first embodiment. FIG. 16 is a diagram showing an example of a computer that executes an analysis program.

Hereinafter, embodiments of the analyzer, analysis method, analysis program, analysis system, and terminal apparatus according to the present application will be described in detail with reference to the drawings. The present invention is not limited to the embodiments described below.

[Structure of the first embodiment]
First, the configuration of the analysis system according to the first embodiment will be described with reference to FIG. FIG. 1 is a diagram showing a configuration example of an analysis system according to the first embodiment. As shown in FIG. 1, the analysis system 1 has a server 10, a client terminal 31, and a client terminal 32. The server 10 is connected to the client terminal 31 and the client terminal 32 via the network 2. The server 10 is an example of an analyzer.

The analysis system 1 is a system for crisis response. Crisis includes natural disasters such as earthquakes, typhoons and tsunamis. The analysis system 1 can support the formulation of a disaster recovery plan.

For example, the analysis system 1 is used in a disaster countermeasures headquarters established by a local government or the like. The disaster response headquarters is an organization that formulates recovery plans for crisis response and issues instructions and orders to the site.

The client terminal is installed at the site related to the restoration work in response to instructions from the disaster countermeasures headquarters or the disaster countermeasures headquarters. For example, an instruction and an instruction are issued as a message from the client terminal installed in the disaster countermeasures headquarters to the task management unit 122 of the server 10. On the other hand, in response to the drafted message, the client terminal installed at the site sends a message to the task management unit 122 as a site report.

Further, each screen generated by the server 10 is displayed on a display device provided in each client terminal. In other words, the server 10 can perform display control on the display device of the client terminal.

As shown in FIG. 1, the server 10 has a storage unit 11 and a control unit 12. The storage unit 11 has a business plan storage unit 111, a task storage unit 112, a topic storage unit 113, a positive / negative model storage unit 114, and a risk detection condition storage unit 115. Further, the control unit 12 has a business plan management unit 121, a task management unit 122, a topic collation unit 123, a positive / negative determination unit 124, and a trend analysis unit 125.

The business plan management unit 121 reads the business plan (checklist) from the business plan storage unit 111, manages the progress of the checklist based on the progress of the tasks associated with the items in the checklist, and sets the checklist as the business plan. It is stored in the storage unit 111. It is assumed that the business plan is created in advance by the user and stored in the business plan storage unit 111.

The business plan management unit 121 manages the matters to be implemented (hereinafter referred to as ToDo) in a checklist format by dividing them into crisis response phases. Further, the business plan management unit 121 can generate a business plan management screen for displaying information about ToDo and display it on each client terminal.

FIG. 2 is a diagram showing an example of a business plan screen. As shown in FIG. 2, the business plan management unit 121 displays the No. 1 on the business plan screen. , Completion status, ToDo list, corresponding organization, completed task, remarks, new task creation button, etc. are displayed. A plurality of tasks are created from each ToDo. The task corresponds to the detailed work to achieve ToDo. The business plan management unit 121 displays the breakdown of the task status in the column of completed tasks.

For example, the business plan management unit 121 sets the display value of the item of the completed task of ToDo in which all the tasks are completed to "completed". Further, in order to realize flexible correspondence, the display value of the item of the completed task may be set manually.

Further, the analysis process (process by the topic collation unit 123, the positive / negative determination unit 124, and the trend analysis unit 125) is performed by the server 10 with the report from each organization as a trigger for the task created for the status report and the summary ToDo. It may be done.

The task management unit 122 manages the priority and progress status of the task by the flags (correspondence status, priority, due date) given to the task drafted in association with the items in the checklist, and manages the task in the task storage unit 112. Store in.

The task management unit 122 manages the priority and progress status of the task by the flags (correspondence status, priority, due date) given to the task drafted in association with the items of the checklist (ToDo). FIG. 3 is a diagram showing an example of a task management screen. As shown in FIG. 3, the task has items such as task ID, priority, status, completion schedule, task name, related task, drafting date and time, sender, destination, contact content, and the like. In particular, the contents of the communication include orders, instructions, reports, etc. related to crisis response.

When a message such as a report is exchanged for the task after the task is created, it can be linked to the task and managed as a thread. This thread is also accumulated in the task storage unit 112. In addition, the task priority flag can be set to 4 levels (urgent and important / important / urgent / normal), and the status flag can be set to 4 levels (not supported / supported / completed / well known), and can be changed according to the task status as appropriate. Is possible.

The topic collation unit 123 extracts the difference between the texts written by the free description of the report on the disaster at each of the plurality of dates and times. The topic collation unit 123 divides the text into topics, and extracts differences between sentences of different texts included in the same topic and having a similarity of a predetermined value or more. For example, the topic collation unit 123 aligns topics between reports whose dates and times are before and after, and stores them in the topic storage unit 113 as topic markers. Then, the topic collation unit 123 extracts the difference in each topic marker unit.

The processing of the topic collation unit 123 will be described with reference to the figure. First, FIG. 4 shows an example of a text in which a report on a disaster at each of a plurality of dates and times targeted by the topic collation unit 123 is written by free description. FIG. 4 is a diagram showing an example of a text in which a report is written.

In FIG. 4, the text containing the report of 2018/7/8 11:38:00 and the text containing the report of 2018/7/8 19:41:00 are shown side by side. In addition, the text has titles such as "building isolation", "building submergence", and "building isolation elimination" in the text. The underline, double underline, dotted line underline, strikethrough, change of character color, and box of rectangle in FIG. 4 do not exist in the original text and are displayed for explanation. Also, the square box represents the topic. Further, the display mode of the text is not limited to that of FIG. For example, the single underline, the double underline, and the dotted underline may be represented by red letters, green letters, and gray letters, respectively.

The topic collation unit 123 first divides the text into topics. FIG. 5 is a diagram showing an example of a topic division method. In the example of FIG. 5, the topic collation unit 123 recognizes "equipment damage" and "repair completion" as titles, and divides the topic by the titles.

The topic collation unit 123 recognizes the titles appearing in the text of the report according to past cases and specific formats. Divide the report by making the topic from the title to the next title. For example, the topic collation unit 123 can recognize a portion to which a symbol character (“□”, “●”, etc.) as a marker is added to the beginning of a sentence as a specific format as a title. For past cases and formats for topic division, initial values are set manually, and registration may be added regularly or at any time when data accumulation progresses and new patterns increase.

Further, the topic collation unit 123 combines the sentences in the topic. Further, the topic collation unit 123 extracts if there is a difference between the combined sentences. FIG. 6 is a diagram showing an example of a method of combining sentences.

As shown in FIG. 6, the topic collation unit 123 combines sentences having a higher degree of similarity within the same topic. For example, the topic collation unit 123 calculates the distance between sentences as the degree of similarity, and obtains a combination from which the total distance is the smallest among the possible combinations. The topic collation unit 123 may acquire all or a part of the combinations such that the total distance is the minimum. Here, the topic collation unit 123 uses a function for calculating the distance between sentences to consider factors such as the type of word optimized by using past data and the position in the sentence.

The topic collation unit 123 uses the function f to calculate the distance, which is the distance between the sentence s1 and the sentence s2, as shown below. Here, the procedure for calculating the distance from all possible combinations is shown, but the topic collation unit 123 may omit a part due to the convenience of processing time or the like.
s1, s2: 2 sentences to be calculated or 2 sentences of the topic
f: Function of distance optimized using known data (calculation between sentences and calculation between sentences are different)
m [0, 0] = 0
for i ← 1 to | s1 |
m [i, 0] = f (s1, s2, i-1, 0, i, 0)
for j ← 1 to | s2 |
m [0, j] = f (s1, s2, 0, j-1, 0, j)
for i ← 1 to | s1 |
for j ← 1 to | s2 |
m [i, j] = min (m [i-1, j-1] + f (s1, s2, i-1, j-1, i, j),
m [i-1, j] + f (s1, s2, i-1, j, i, j),
m [i, j-1] + f (s1, s2, i, j-1, i, j))

The positive / negative determination unit 124 determines whether the difference extracted by the topic collation unit 123 indicates deterioration of the situation or improvement of the situation. For example, the positive / negative determination unit 124, when the difference is due to a change in quantity, indicates whether the difference indicates an exacerbation of the situation or a situation based on whether the change is an increase or decrease in quantity. Judge whether it indicates an improvement in.

First, the positive / negative determination unit 124 reads the topic marker between reports having different dates and times from the topic storage unit 113, and extracts the difference under the topic. Further, the positive / negative determination unit 124 applies the positive / negative determination model read from the positive / negative model storage unit 114 to the extracted difference to perform positive / negative determination. Then, the positive / negative determination unit 124 stores the determination result in the topic storage unit 113 and displays it on each client terminal.

Specifically, the positive / negative determination unit 124 first acquires a matching portion between sentences and discriminates between an unchanged portion and a changed portion. After that, the positive / negative determination unit 124 determines whether the change indicates deterioration or improvement of the situation based on the change in quantity or expression of the changed portion.

The initial positive / negative determination unit 124 is based on a manual evaluation. In this case, data for manually determining whether or not the changed portion indicates deterioration or improvement is manually prepared, and a positive / negative determination model is generated from the pattern based on the data. Regular or occasional registrations are added to the positive / negative determination model when data accumulation progresses and new patterns increase.

For example, the positive / negative judgment model is an evaluator created from a manual evaluation for each topic as to whether the increase in quantity indicates deterioration or improvement in the expression including quantity. be. The positive / negative judgment model accumulates cases showing deterioration in expression and cases showing improvement, and uses them for judgment.

On the other hand, even in the changed part, there may be cases where the change in quantity is not considered to indicate deterioration or improvement from the context before and after, and there are cases where it is judged that information not written in the sentence is necessary. Data for determining such data may be manually prepared, and a pattern based on the data may be added to the positive / negative determination model.

First, the positive / negative determination unit 124 determines whether each sentence of the latest report is an unchanged portion or a changed portion. Next, the positive / negative determination unit 124 determines whether the changed portion is a positive indicator of improvement in the situation or a negative indicator of the deterioration of the situation. At this time, the positive / negative determination unit 124 may determine that it is unknown if the positive / negative cannot be determined.

As shown in FIG. 4, the positive / negative determination unit 124 can express the determination result by changing the display mode such as the font, style, and ruled line of the text. The positive / negative determination unit 124 may display a screen with a changed display mode on each client terminal as shown in FIG. Further, as shown in FIG. 10, the positive / negative determination unit 124 may express the difference in topic by the difference in the pattern, and clearly display the version (difference in date and time) in which the positive or negative is determined. FIG. 10 is a diagram that gives a bird's-eye view of changes in each topic.

FIG. 7 is a diagram showing an example of a positive / negative analysis method. In this case, an increase in quantity means an increase in the area where repairs have been completed. In other words, it is conceivable that the manual evaluation data for such an increase in quantity indicates an improvement in the situation. Therefore, the positive / negative determination unit 124 determines that the increase in the quantity is an improvement of the situation from the correction history of the accumulated manual evaluation data.

FIG. 8 is a diagram showing an example of a positive / negative analysis method. In this case, the quantity is a number for identifying which report it is, and does not directly mean improvement or deterioration of the situation. Therefore, the positive / negative determination unit 124 does not determine whether the increase in the quantity is an improvement or a deterioration of the situation from the correction history of the accumulated manual evaluation data, and determines, for example, that it is unchanged.

The trend analysis unit 125 reads the determination result of the positive / negative determination from the topic storage unit 113, applies the risk detection condition read from the risk detection condition storage unit 115, and extracts the part where the risk is suspected. The trend analysis unit 125 displays the extracted portion in a graph, table, or the like and stores it in the risk detection condition storage unit 115. Further, the trend analysis unit 125 may notify the task management unit 122 of the extracted portion as a task.

For example, the trend analysis unit 125 applies the risk detection condition table read from the risk detection condition storage unit 115 to the positive / negative determination result, extracts the part where the risk is suspected, visualizes it with a graph, a table, or the like, and the user. And store it in the risk detection condition storage unit 115.

FIG. 9 is a diagram showing an example of a risk detection table. The "red character" in FIG. 9 corresponds to the red character in FIG. 4, and means the character of the sentence indicating that the situation has worsened. Further, the trend analysis unit 125 can represent the analysis result using the risk detection table in a graph as shown in FIG. FIG. 11 is a diagram showing an example of a graph showing the result of comparing texts. As a result, the user can intuitively grasp the deterioration and improvement of the situation according to the change in the reporting date and time.

Further, the trend analysis unit 125 may display a graph as shown in FIG. 11 on the client terminal. As described above, the trend analysis unit 125 has different modes of the information regarding the difference determined by the positive / negative determination unit 124 to indicate the deterioration of the situation and the difference determined by the positive / negative determination unit 124 to indicate the improvement of the situation. It can be displayed on the display device with.

The trend analysis unit 125 may raise a risk detection result as a task and notify the user. In that case, the trend analysis unit 125 creates a new task, issues it as a thread as a response to the original reporting task, or associates it with the "related task" of the original reporting task. be able to.

In addition, the trend analysis unit 125 may perform analysis and screen display of the analysis results as shown in FIGS. 12, 13 and 14. 12, 13 and 14 are diagrams showing examples of the results of comparing texts.

For example, the graph of FIG. 12 shows the breakdown of the number of characters of the sentences determined to be improved, deteriorated, already appearing, and first appearing in the text having the date and time in the context. The first appearance is a sentence that does not have a sentence to be combined with the past reports. In addition, the above-mentioned sentences are sentences that have a combination sentence in the past reports but have not changed.

The graph on the left side of FIG. 13 shows the breakdown of the number of characters in the sentence determined to be improved, deteriorated, already appeared, or first appeared in the text containing the latest report. The graph on the right side of FIG. 13 shows the breakdown of the number of characters in the sentence determined to be improved or deteriorated. Further, the graph of FIG. 14 shows changes in the breakdown of the number of characters in the sentences determined to be improved, deteriorated, already appeared, and first appeared in the plurality of texts.

[Processing of the first embodiment]
FIG. 15 is a diagram showing a processing flow of the analysis system according to the first embodiment. As shown in FIG. 15, the business plan management unit 121 sets a checklist (step S101). The business plan management unit 121 creates a task from the checklist (step S102). Then, the business plan management unit 121 requests the task management unit 122 to register the task (step S103).

The task management unit 122 registers the task (step S104). Then, the task management unit 122 requests the topic collation unit 123 to collate the topic (step S105). Here, the topic collation unit 123 collates the topic (step S106). The topic collation unit 123 requests the positive / negative determination unit 124 to make a positive / negative determination (step S107).

The positive / negative determination unit 124 makes a positive / negative determination (step S108). Then, the visualization of the trend is requested (step S109). Visualization of the trend here includes generating a graph as shown in FIG. 11, for example. The trend analysis unit 125 analyzes the trend (step S110). At this time, the trend analysis unit 125 may generate a graph.

The task management unit 122 registers a task at the request of the trend analysis unit 125 (step S111). Further, the task management unit 122 manages the progress of the task (step S112). Then, the task management unit 122 notifies the business plan management unit 121 of the task progress (step S113). Further, the series of processes resulting from step S102 are repeatedly executed while the crisis response is being implemented.

[Effect of the first embodiment]
As described above, the topic collation unit 123 extracts the difference between the texts in which the report on the disaster at each of the plurality of dates and times is written by free description. The positive / negative determination unit 124 determines whether the difference extracted by the topic collation unit 123 indicates deterioration of the situation or improvement of the situation. In this way, the analysis system 1 can extract a part of interest from the user from the freely described text. As a result, according to the present embodiment, it is possible to efficiently grasp the contents of the report in disaster countermeasures.

The topic collation unit 123 divides the text into topics, and extracts differences between sentences of different texts included in the same topic and having a similarity of a predetermined value or more. This makes it possible to effectively extract differences by grasping the contents of the report.

When the difference is due to a change in quantity, the positive / negative determination unit 124 determines whether the difference indicates an exacerbation of the situation or improves the situation based on whether the change is an increase or decrease in the quantity. Is determined. This makes it possible to accurately grasp the content of the report based on the objective information of changes in quantity.

The trend analysis unit 125 displays the information regarding the difference determined by the positive / negative determination unit 124 to indicate the deterioration of the situation and the difference determined by the positive / negative determination unit 124 to indicate the improvement of the situation in different modes from each other. Display. As a result, the user can intuitively grasp the deterioration, improvement, etc. of the situation between reports with different dates and times.

[System configuration, etc.]
Further, each component of each of the illustrated devices is a functional concept, and does not necessarily have to be physically configured as shown in the figure. That is, the specific forms of distribution and integration of each device are not limited to those shown in the figure, and all or part of them may be functionally or physically dispersed or physically distributed in arbitrary units according to various loads and usage conditions. Can be integrated and configured. Further, each processing function performed by each device is realized by a CPU (Central Processing Unit) and a program that is analyzed and executed by the CPU, or hardware by wired logic. Can be realized as.

Further, among the processes described in the present embodiment, all or part of the processes described as being automatically performed can be manually performed, or the processes described as being manually performed can be performed. All or part of it can be done automatically by a known method. In addition, the processing procedure, control procedure, specific name, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified.

[program]
In one embodiment, the server 10 can be implemented by installing an analysis program that executes the above processing as package software or online software on a desired computer. For example, by causing the information processing apparatus to execute the above analysis program, the information processing apparatus can function as the server 10. The information processing device referred to here includes a desktop type or notebook type personal computer. In addition, information processing devices include smartphones, mobile communication terminals such as mobile phones and PHS (Personal Handyphone System), and slate terminals such as PDAs (Personal Digital Assistants).

Further, the server 10 may be implemented as a device in which the terminal device used by the user is a client and the service related to the above processing is provided to the client. For example, the server 10 is implemented as a device that provides an analysis service that inputs a plurality of freely described texts and outputs a determination result of a difference. In this case, the server 10 may be implemented as a Web server, or may be implemented as a cloud that provides services related to the above processing by outsourcing.

FIG. 16 is a diagram showing an example of a computer that executes an analysis program. The computer 1000 has, for example, a memory 1010 and a CPU 1020. The computer 1000 also has a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. Each of these parts is connected by a bus 1080.

The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM (Random Access Memory) 1012. The ROM 1011 stores, for example, a boot program such as a BIOS (BASIC Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. For example, a removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to, for example, a mouse 1110 and a keyboard 1120. The video adapter 1060 is connected to, for example, the display 1130.

The hard disk drive 1090 stores, for example, the OS 1091, the application program 1092, the program module 1093, and the program data 1094. That is, the program that defines each process of the server 10 is implemented as a program module 1093 in which a code that can be executed by a computer is described. The program module 1093 is stored in, for example, the hard disk drive 1090. For example, the program module 1093 for executing the same processing as the functional configuration in the server 10 is stored in the hard disk drive 1090. The hard disk drive 1090 may be replaced by an SSD (Solid State Drive).

Further, the setting data used in the processing of the above-described embodiment is stored as program data 1094 in, for example, a memory 1010 or a hard disk drive 1090. Then, the CPU 1020 reads the program module 1093 and the program data 1094 stored in the memory 1010 and the hard disk drive 1090 into the RAM 1012 as needed, and executes the process of the above-described embodiment.

The program module 1093 and the program data 1094 are not limited to those stored in the hard disk drive 1090, and may be stored in, for example, a removable storage medium and read out by the CPU 1020 via the disk drive 1100 or the like. Alternatively, the program module 1093 and the program data 1094 may be stored in another computer connected via a network (LAN (Local Area Network), WAN (Wide Area Network), etc.). Then, the program module 1093 and the program data 1094 may be read from another computer by the CPU 1020 via the network interface 1070.

1 Analysis system 2 Network 10 Server 11 Storage unit 12 Control unit 31, 32 Client terminal 111 Business plan storage unit 112 Task storage unit 113 Topic storage unit 114 Positive negative model storage unit 115 Risk detection condition storage unit 121 Business plan management department 122 Task management Department 123 Topic collation unit 124 Positive negative judgment unit 125 Trend analysis unit

Claims (8)

An extractor that extracts the differences between texts written by free description for reports on disasters at each of multiple dates and times,
A determination unit for determining whether the difference extracted by the extraction unit indicates deterioration of the situation or improvement of the situation.
An analyzer characterized by having. The extraction unit according to claim 1, wherein the text is divided into topics, and differences between sentences of different texts included in the same topic and having a similarity of a predetermined value or more are extracted. Described analyzer. When the difference is due to a change in quantity, the determination unit determines whether the difference indicates an exacerbation of the situation or a situation based on whether the change is an increase or decrease in quantity. The analyzer according to claim 1 or 2, wherein it determines whether or not it indicates improvement. Further having a display control unit for displaying information on the difference determined by the determination unit to indicate deterioration of the situation and the difference determined by the determination unit to indicate improvement of the situation on the display device in different modes from each other. The analyzer according to any one of claims 1 to 3, wherein the analyzer is characterized by the above. An analytical method performed by an analyzer,
An extraction process that extracts the differences between texts written by free description for reports on disasters at each of multiple dates and times, and
A determination step for determining whether the difference extracted in the extraction step indicates deterioration of the situation or improvement of the situation.
An analysis method characterized by including. An analysis program for operating a computer as the analyzer according to any one of claims 1 to 4. An analysis system that has a server and a terminal device.
The server
An extractor that extracts the differences between texts written by free description for reports on disasters at each of multiple dates and times,
A determination unit for determining whether the difference extracted by the extraction unit indicates deterioration of the situation or improvement of the situation.
Have,
The terminal device is
It is characterized by having a display unit that displays information on a difference determined by the determination unit to indicate deterioration of the situation and a difference determined by the determination unit to indicate improvement of the situation in different modes. Analysis system. Based on the judgment result that the report on the disaster at each of the multiple dates and times determines whether the difference between the texts written by the free description indicates the worsening of the situation or the improvement of the situation. The terminal device is characterized by having a display unit that displays information about a difference determined to indicate a deterioration of the situation and a difference determined to indicate an improvement of the situation in different modes from each other.

Images