JP5972425B1 - Reputation damage risk report creation system, program and method - Google Patents

Abstract

An object of the present invention is to create a reputation damage report that quantifies and presents the degree of influence of reputation on society depending on the media and posting location of reputation expressions. A reputation damage risk report creation system according to the present invention allows a user to search a large number of web pages on the Internet by inputting a specific keyword by each of a plurality of search programs, and performs a search at a place where search results are extracted. Identify the information provision medium and the publication location where the risky expression that appears with the keyword is found, determine the influence index value of the information provision medium providing the search program and the influence index value of the publication location of the search results, A total of these influence index values is obtained as a reputation damage risk degree, and a reputation damage risk report 20 is created. [Selection] Figure 4

Description

  The present invention relates to a reputation damage risk report creation system, program, and method for searching a reputation expression of a company on the Internet web and automatically creating a reputation damage risk report.

  In order to praise the corporate image, others may deliberately advertise on the Internet web. In recent years, a reputation monitoring service that monitors web pages where such reputation expressions are seen, creates reputation damage reports, and distributes them to requesters has become widespread.

  In such a reputation monitoring service, a search engine provided on a major portal site is used to perform a wide search using the requesting company name as a keyword, and reputation expressions registered in the dictionary in advance are shared with the company name. The web page that is displayed is extracted, analyzed what kind of reputation expression appears at what frequency, and distributed to the requesting company as a reputation damage report.

  In addition, at the requesting company that received the reputation damage report, the person in charge looks at the reputation damage report and determines whether or not the reputation damage countermeasures are necessary depending on whether the corporate image may be significantly impaired. Ask the company to take measures for reputational damage.

  However, conventionally, a web page where a reputation expression can be seen together with a company name of a requester is extracted without omission and a reputation damage report is created. Therefore, the search keyword suggestion display frame (also called input candidate display frame, input auxiliary function, etc.) based on the search result of the search engine of which portal site, the page number of the search result, and the search keyword. The influence of reputation on society varies depending on the media and posting location of the reputation expression, such as whether it is also displayed in the display column or the related search word is also displayed in the display column, etc. Reputation damage reports cannot be created automatically.

  For example, Patent Document 1 discloses a technique for weighting each reputation expression and automatically reporting a URL that lists the reputation expression with a high score. However, there is no disclosure of a technique for calculating the risk level based on the type of URL or page number in which the same reputation expression is posted.

  Patent Document 2 discloses a technique for monitoring a web page on which bad reputations are posted and creating a report. However, there is no disclosure of a technique for calculating the risk level of reputation damage based on the type or URL of the URL where the same reputation expression is posted.

  Patent Document 3 discloses a technique for collecting and analyzing good evaluations and bad evaluations for a certain product based on the reputation of a website. However, a technique for calculating the risk level of reputational damage is not disclosed.

  Patent Document 4 discloses a technique for creating a report by assigning a weight to a reputation expression and creating a URL with a high score of bad reputation expression as a caution URL. However, as in Patent Document 1, there is no disclosure of a technique for calculating the degree of risk based on the type of URL, page number, or the like that uses the same reputation expression.

  Further, Patent Document 5 discloses a technique for extracting a URL having a reputation expression and creating a report. However, there is no disclosure of a technique for calculating the risk level of reputation damage based on the type or URL of the URL where the same reputation expression is posted.

JP 2003-271609 A JP 2003-303242 A Japanese Patent Laid-Open No. 2004-070405 Japanese Patent Laid-Open No. 2004-280568 Japanese Patent Laying-Open No. 2005-063242

  The present invention has been made in view of the above-mentioned problems of the prior art, and calculates the impact index value of reputation damage that varies depending on the type and page number of the web page on which the same reputation expression is posted, and It is characterized by providing a reputation damage risk report creation system, program, and method capable of creating a reputation damage report that quantifies and presents the degree of reputation impact on society according to the medium and the posting location of the expression.

  One feature of the present invention is that an influence index value of each of a plurality of information providing media providing information on a computer screen and an influence index value of a place where a search keyword is placed in each of the plurality of information providing media, Retrieval that is extracted by searching for information on the Internet by inputting a specific keyword and searching for information on the Internet by entering an influence degree index value storage part that holds each as an influence degree index value, a risk word storage part that holds a plurality of risk words A risk-containing expression search unit for identifying a posted part in which a risk-containing expression in which any of a plurality of risk words held by the risk word storage unit appears together with the input search keyword is found , The impact index value storage for the location of the risk expression specified by the risk expression search unit The impact index value of the information providing medium that provides the information and the impact index value of the place where the search result is posted are calculated, and the total of these impact index values is calculated as the reputation damage risk level. Reputation damage risk degree calculation unit obtained as, and the input search keyword and extracted risk word, the reputation damage risk degree obtained by the reputation damage risk degree calculation unit, information related to the extracted information providing medium, and A reputation damage risk report creation system comprising a reputation damage risk report creation section that creates and outputs a reputation damage risk report including information related to the posting location of the risk expression.

  Another feature of the present invention is that a search step of inputting a specific keyword and searching for information on the Internet by a search program, an information providing medium providing a search result extracted by the search step, and its posting location A risk-containing expression search step for identifying a posting location where a risk-containing expression in which any of the plurality of risk words held by the risk word storage unit appears together with the input search keyword is found, and the risk-containing expression The impact index value of each of the plurality of information providing media held in the impact index value storage unit and the influence of the search keyword posting location on each of the plurality of information providing media for the place of the risk-containing expression identified by the search step Refer to the data of the degree index value, and the influence degree index value of the information providing medium Reputation index value of the posted part of the search result is calculated, and a reputation damage risk degree calculation step for obtaining the total of these influence index values as a reputation damage risk degree, the input search keyword and the extracted risk word, Create and output a reputation damage risk report including the reputation damage risk degree obtained in the reputation damage risk degree calculation step, information related to the extracted information providing medium, and information related to the posting location of the risk expression included A reputation damage risk report creation program for causing a computer to execute a reputation damage risk report creation step.

  Still another feature of the present invention is that a computer is used to input a specific keyword and a search step for searching information on the Internet by a search program, and information providing a search result extracted by the search step A risk-containing expression search step for identifying a posting location in which a risk-containing expression in which any of a plurality of risk words held by the risk word storage unit appears together with the input search keyword is found in the provided medium and its posting location. And a search in each of the plurality of information providing media stored in the influence index value storage unit and each of the plurality of information providing media for the place of the risk containing expression specified by the risk containing expression search step Refer to the data with the impact index value of the keyword placement, and the information Reputation damage risk degree calculation step of determining the influence index value of the provided medium and the influence index value of the place where the search results are posted, and obtaining the total of these influence index values as the reputation damage risk degree, and the input The search keyword and the extracted risk word, the reputation damage risk degree obtained in the reputation damage risk degree calculation step, the information related to the extracted information providing medium, and the information related to the posted location of the risk expression are included. A reputation damage risk report creating method for automatically creating a reputation damage risk report by executing a reputation damage risk report creating step of creating and outputting a reputation damage risk report.

  According to the present invention, the type of information providing medium such as a portal site, bulletin board site, blog site, provider site, etc. in which the same reputation expression is posted, the page number being posted, and the display in the input candidate display frame Calculate the impact index value of reputational damage that varies depending on the location, such as the position and display order in the search candidate word column, and quantify the impact of reputation on society depending on the media and posting location of the reputation expression. A reputational damage report can be created.

The block diagram of the function structure of the reputation damage risk report creation system of 1st Embodiment. Explanatory drawing of the data table for reputation damage risk index value calculation used with the reputation damage risk report creation system of 1st Embodiment. The flowchart of the reputation damage risk report creation process which the reputation damage risk report creation system of 1st Embodiment performs. Explanatory drawing of the reputation damage risk report created by the reputation damage risk report creation system of 1st Embodiment. The block diagram of the function structure of the reputation damage risk report production system of 2nd Embodiment. The first half of the flowchart of the reputation damage risk report creation processing executed by the reputation damage risk report creation system according to the second embodiment. The latter half part of the flowchart of the reputation damage risk report creation process which the reputation damage risk report creation system of 2nd Embodiment performs. The flowchart which shows the procedure which accesses and acquires the reputation damage risk report created by the reputation damage risk report creation system of 2nd Embodiment from a contract company. Explanatory drawing of the reputation damage risk report produced by the reputation damage risk report creation system of 2nd Embodiment. Explanatory drawing of the data table for reputation damage risk index value calculation used with the reputation damage risk report creation system of 2nd Embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
<First Embodiment>
As shown in FIG. 1, the reputation damage risk report creation system of the embodiment is a computer system 1 that is connected to the Internet 100 and executes various processing functions described below. The reputation damage risk report creation system according to the present embodiment may be composed of a distributed computer system or a stand-alone computer system. In the embodiment, a computer system 1 is shown. It is assumed that a user interface necessary for data input, operation instruction input, information display, and information output is naturally connected to the computer system 1 such as a display, a keyboard, and a pointing device. Various software programs necessary for realizing the following embodiments are installed, and a large-capacity storage device 2 such as a hard disk drive is incorporated to store a large amount of data required for the system. Or it shall be provided externally. Reference numerals 1 </ b> A, 1 </ b> B, and 1 </ b> C are contractor computer systems connected to the Internet 100.

  The functional configuration of the reputation damage risk report creation system according to the present embodiment is as follows, and each of the plurality of information providing media providing information on the computer screen and each of the plurality of information providing media 2 holds an influence index value storage unit 11 that holds an influence index value of a search keyword posting location in the form of a data table 10 as shown in FIG. 2, and holds a number of risk words set in advance as risk words. Risk word storage unit 12 and keyword registration unit 13 holding keywords of a reputation damage search target that requires a reputation damage risk report, such as a company name, a store name, a product name, a trademark, etc. It has.

  The reputation damage risk report creation system according to the present embodiment allows a search program such as various search engines provided on the Internet to specify a specific search keyword and search a large number of web pages on the Internet without omission. Included in the posted location of the extracted search result is an identification that identifies a posted location where any of the plurality of risk words held by the risk word storage unit 13 is found along with the input search keyword. The risk expression search unit 14 and the influence of the information providing medium that provides the information by referring to the data of the influence index value storage unit 11 for the place of the risk expression included by the risk expression search unit 14 Index value and the impact index value of the place where search results are posted, and the total of these impact index values Reputation damage risk degree calculation unit 15 obtained as a reputation damage risk degree, input search keyword and extracted risk word, reputation damage risk degree obtained by the reputation damage risk degree calculation unit 15, and the extracted information providing medium And a reputation damage risk report creation section 16 that creates and outputs a reputation damage risk report 20 including information related to the information to be posted and information related to the place where risk expression is included.

  When the risk-containing expression search unit 14 finds the risk-containing expressions for the specified search keyword at a plurality of locations, the reputation damage risk report creation unit 16 arranges the risk-containing expressions at the plurality of locations in descending order of the reputation damage risk level. The reputation damage risk report 20 is created and output.

  As shown in FIG. 2, in the data table 10 held by the influence index value storage unit 11, a large number of information providing media providing search programs such as a search engine and a search API as the influence index values, for example, For each of the portal site, provider site, news providing site, bulletin board site, blog site, etc., the score and weight are determined for each information providing medium. Further, for the search result posting page, a score and a weight are determined for each page on which the risk expression is posted. Further, for the display in the A column that displays the input candidates for the designated search keyword, the score and the weight are determined for each display line in the A column. As for the display in column B for displaying related search keywords, a score and a weight are determined for each display order. In addition, for the C column such as the organic display column for displaying the natural language, the score and the weight are determined for each display order, and the display in any one of the D columns which are the display columns of a plurality of special modules is performed for each special module. The score and weight are determined.

  For example, when the domain of the posting site is a well-known “abcd1.co.jp”, a risk word appears together with a keyword on the first page of the posting page of the search result [medium score = 100, weight = 0.3] is determined and registered, and at the same time, [page score of posted page = 100, weight = 0.5] is determined and registered. When the domain of a publishing site that is highly used as a search engine is “abcd2.co.jp”, it is determined and registered in substantially the same manner.

  In addition, the domain abcd1. co. When a search keyword is input using the search engine of jp, some other keywords co-occurring are displayed together with the search keyword in the A column which is an input candidate display frame. When the risk word is displayed together with the search keyword at the first place in the display, it is determined and registered as [publication place score = 100, weight = 0.8]. Also, the same domain abcd1. co. When the search keyword is input using the search engine of jp, if it appears first in the B column displaying the related search word, it is determined that [publication place score = 95, weight = 0.8]. And register. Furthermore, as illustrated in the data table 10 of FIG. 2, when the first place in the column C, which is a natural language search result, appears [placement score = 95, weight = 0.8], the special module If it is displayed in the first place in the D column, it is determined and registered as [publication place score = 95, weight = 0.8] and the like.

  The reputation damage risk level calculation unit 15 refers to the data table 10 of FIG. 2 according to which posting location of the risk-containing expression is in which information providing medium, and calculates the product of each score and weight. Calculate the total and calculate the reputational damage risk level.

  Next, the processing operation of the reputation damage risk report creation system configured as described above will be described with reference to the flowchart of FIG. At the same time, this is a method for creating a reputation damage risk report according to an embodiment of the present invention, which is implemented by a computer, and is realized by installing and executing a corresponding reputation damage risk report creation program in a computer system. is there.

  Set and save data necessary for initial settings, new contracts, system maintenance, etc. In that case, register various keywords that are likely to be damaged if the reputation is made on the net, such as the contractor's company name, trademark, trade name, product name, etc., and various bad reputation expressions, A risk word is registered (step S1).

  In addition, the degree of adverse social impact caused by the occurrence of the above-mentioned keywords and risk words co-occurring in some place on the search screen is used as an influence index value based on the posting page, posting location and weight. The set data table 10 of FIG. 2 is also created (step 5). The process before and after step S1 and step S5 is not particularly determined. In addition, new registration, additional registration, and revision of keywords are repeatedly performed at any timing as necessary.

  After the registration of these various data, for example, the risk expression search is executed regularly every day, every week, every month, or at any timing as required (step S11). In this risk-containing expression search, the risk-containing expression search unit 14 causes the Yahoo. co. jp and Google. co. jp and bing. co. Use a search engine provided on a portal site on the Internet, such as jp, and specify one or more types of keywords registered in the keyword registration unit 13 to leak a large number of web pages on the Internet Let's search without. In addition, a search program such as a search engine or a search API prepared for a provider site provided by a provider, a news providing site, a blog site, a bulletin board site, or the like on the information providing medium site. Use to search. Then, in a large number of extracted web pages, one of a plurality of risk words held by the risk word storage unit 12 is specified as a place where a risk-containing expression appears along with a designated search keyword. In the description of the present embodiment, the portal site domains that are frequently used are exemplified as “abcd1.co.jp” and “abcd2.co.jp”. However, it is not limited to the main portal site, and provider sites, news providing sites, blog sites, bulletin board sites and the like as exemplified above are also included in the search target.

  Next, the reputation damage risk degree calculation unit 15 refers to the data table 10 of the influence index value storage unit 11 and provides a search engine for each of the posting locations of the many risky expressions specified by the risk expression expression search unit 14. The influence index value of the portal site being operated and the influence index value of the place where the search results are posted are calculated, and the total of these influence index values is obtained as the reputation damage risk degree (step S15).

Now, the search keyword is “ABC Shoji Co., Ltd.”, and the portal site abcd1. co. When a search is performed by a search engine on jp, the following occurs.
About the search keyword “ABC Shoji Co., Ltd.”, abcd1. co. When “ABC Shoji Co., Ltd.” was entered into the search window using the search engine on jp, “ABC Shoji Co., Ltd. Black Company” was displayed in the third place in the A column.

When a search is performed using “ABC Shoji Co., Ltd.” as a keyword, a large number of web pages including “ABC Shoji Co., Ltd.” are extracted, and the search results are divided into two pages. The web pages were also extracted at the same time, and the URL links of those web pages were displayed with a few lines of excerpts.
Furthermore, in the B column on the first page of the search result, the notation “ABC Shoji Co., Ltd. Black Company” was found fifth. In addition, for each web page of the search result, the web page including the risk word “black company” is the first page of the search result, the second and tenth positions in the C column, and the fifth page in the C column. Was found.

  If the above is the search result, the reputation damage risk level is calculated as follows. Reputation damage risk level = (medium score × weight + page score × weight + placement location score × weight). When this formula is applied to the above example, the media score is the famous portal site abcd1. co. Since it is jp, the media score of the posting site domain is 100 × weight 0.3 = 30. For the first page of the search results, the posted page score is 100 × weight 0.5 = 50, and the second page of the C column is the posted location score 85 × weight 0.8 = 68, C For the web page in the 10th place in the column, the posted part score is 5 × weight 0.8 = 4.

Further, for the web page in the fifth column of column C in the second page of search results, the media score of the posted site domain is 100 × weight 0.3 = 30, and the page score of search results is 20 × weight 0.5. = 10, placement location score 60 × weight 0.8 = 48.
And since it is the 3rd place in the A column, the media score of the posting site domain 100 × weight 0.3 = 30, the posting page of the search result is the first page, the posting page score 100 × weight 0.5 = 50, posting Location score 80 × weight 0.8 = 64.

  From such a search result, when the reputation monitoring requesting company is “ABC Shoji Co., Ltd.”, a reputation damage risk report 20 as shown in FIG. 4 is created. The first risk degree is abcd1. co. This is a search result that appears in the second position of the first page and column C in the search of the search site of jp, and has a risk degree of 148 (= 30 + 50 + 68). The second highest risk degree is abcd1. co. This is a search result that appears in the third position of the first page in the search of the search site of jp, and has a risk degree of 144 (= 30 + 50 + 64).

  For the reputation damage risk report automatically created in this way, the form is further processed to create a reputation damage risk report for each search keyword, or multiple risk words are grouped for each search keyword, You can create a reputational damage risk report that has been rearranged in descending order.

  In the risk-containing expression search in step S11 of FIG. 3, in order to increase the effectiveness of the keyword search, the above search keywords “ABC Trading Co.”, “ABC Trading”, “ABC Inc.”, “ ○ Derived keywords such as “BC Trading Co., Ltd.”, “A ○ C Shoji”, etc., Omission of characters such as “Corporation”, etc. Generates fluctuations such as obscure characters, hit characters, conversion errors, typographical errors, etc. In addition, the reliability of the reputation damage risk report is improved by searching for a web page that co-occurs with the risk word.

<Second Embodiment>
Next, a reputation damage risk report creation system, a reputation damage risk report creation method, and a reputation damage risk report creation program according to another embodiment of the present invention will be described with reference to FIGS. The reputation damage risk report creation method of this embodiment is also realized by a computer by installing and executing a reputation damage risk report creation program corresponding to the computer system.

  FIG. 5 shows a functional configuration of the reputation damage risk report creation system of the present embodiment. The computer system 101 used is the same as that in the first embodiment. The reputation damage risk report creation method and program according to the present embodiment perform more detailed processing than the first embodiment. In FIG. 5, reference numerals 101 </ b> A, 101 </ b> B, and 101 </ b> C are contractor computer systems connected to the Internet 100.

  The reputation damage risk report creation system according to the present embodiment includes an influence index value of each of a plurality of information providing media providing a search engine on the screen of the computer 101, and a search result posting location in each of the plurality of information providing media. 9 is an influence index value storage unit 111 that holds the influence index values in the form of the data table 110 as shown in FIG. 9, and a risk word storage unit that holds a number of risk words set in advance as risk words 112, and a keyword storage unit 113 for holding a reputation damage target keyword requiring a reputation damage risk report, such as a company name, a store name, a product name, a trademark, etc. of a customer who has concluded a contract.

  The reputation damage risk report creation system according to the present embodiment is also provided on the Internet by specifying a specific keyword using a search program such as a search engine or a search API provided by each information providing medium provided on the Internet 100. A risk that includes one of a plurality of risk words that the risk word storage unit 112 holds together with a designated search keyword in a search result place where a large number of web pages are searched without omission. A risk-containing expression search unit 114 is provided for specifying a placement where an expression is found.

  As a feature of the present embodiment, in order to increase the effectiveness of the keyword search, the risk-containing expression search unit 114, for example, “ABC Trading”, “ Considering fluctuations such as abbreviations, guesses, conversion mistakes, typo, etc., derived keywords such as “ABC Co., Ltd.”, “○ BC Shoji Co., Ltd.”, “A ○ C Shoji Co., Ltd.”, etc. In addition to the search keyword, they are also searched for a web page that co-occurs with the risk word.

  The reputation damage risk report creation system according to the present embodiment further includes the data table 110 of FIG. 9 held by the influence index value storage unit 111 for the place where the risk expression is included specified by the risk expression search unit 114. Reputation damage risk level calculation by calculating the impact index value for each information providing medium and the impact index value of the search result publication location, and calculating the total of these impact index values as the reputation damage risk level. Section 115, specified search keyword and extracted risk word, reputation damage risk level calculated by reputation damage risk degree calculation section 115, information of information providing medium in which risk expression is included, and posting of risk expression A reputation damage risk report creation unit 116 is provided that creates and outputs a reputation damage risk report including information on the location.

  When the risk expression search unit 114 finds the risk expression for the specified keyword in a plurality of locations, the reputation damage risk report creation unit 116 has a list format in which the risk inclusion expressions in the plurality of locations are arranged in descending order of the reputation damage risk level. A reputation damage risk report 120 as shown in FIG. 10 is created and output.

  In the data table 110 held by the influence index value storage unit 111 shown in FIG. 9, a score and a weight are provided for each of a plurality of information providing media that provide search programs such as a search engine and a search API as influence index values. A score and a weight are set for each page where the relevant risk expression is posted for the search result posting page, and the display in the A column for displaying input candidates for the designated search keyword A score and a weight are set for each display line, a display of a column B displaying a related search word, a score and a weight are set for each display order, and a display of a column C displaying a natural language analysis result is also displayed. A score and a weight are set for each rank, and for display on the special module, a score and a weight are set for each special module in the D column.

  Furthermore, as a feature of the present embodiment, there is a risk word relevance item, and the relevance of the registered risk word and the search keyword is quantified. Any one of up to 0 is set. For example, in the same example as in the first embodiment, the domain of the posting site is abcd1. co. jp, when a risk word appears with a keyword on the first page of the posted page of the search result, [medium score = 100, weight = 0.3] is determined and registered, and at the same time, [page of posted page] Score = 100, weight = 0.5] and register. The domain of another posting site that is highly used as a search engine is abcd2. co. In the case of jp, it is determined in substantially the same manner.

  Also, abcd1. co. When a search keyword is input using the search engine of jp and a risk word is displayed together with the search keyword in the first place in the A column, [publication place score = 100, weight = 0.8] And register. The same abcd1. co. When a search keyword is input using the search engine of jp, if it appears first in the B column, it is determined and registered as [publication place score = 95, weight = 0.8].

  Furthermore, as illustrated in the data table 110 of FIG. 9, when the first place appears in the C column as the organic search result, [publication place score = 95, weight = 0.8], it is a special module. When it is displayed in the first place in the column D, it is determined and registered as [publishing place score = 95, weight = 0.8] or the like.

  Furthermore, the degree of relevance between the search keyword and the risk word is determined depending on how far the risk word is displayed with respect to the search keyword. If the degree of relevance is high, the degree of risk word relevance = 0.9. If the relevance level is medium, the risk word relevance level is set to 0.5. For example, if “ABC Shoji Co., Ltd. is a black company”, almost equal expression, risk word relevance = 1, “ABC trading is listed in the black company list”. In such a case, the risk word relevance is determined to be 0.9.

  The reputation damage risk degree calculation unit 115 refers to the data table 110 in FIG. 9 according to which information provision medium is included in the information provision medium, and calculates the product of each score and weight. Total, and then multiply the risk word relevance as a coefficient to calculate the reputational damage risk.

  Next, the processing operation of the reputation damage risk report creation system configured as described above will be described with reference to the flowcharts of FIGS. The illustration is the same as in the first embodiment, but a more accurate reputation damage risk report is possible than in the first embodiment in that the risk word relevance is incorporated as a coefficient.

Step S101: Register a company name, a product name, a trademark, etc., which are keywords specified by the contractor, keywords proposed to the contractor, and the like, which are to be searched for reputation damage risk.
Step S105: Register various risk words that cause damage to reputation. For example, “black company”, “service overtime”, “hazardous food”, “poisoning”, “corruption” and so on.
Since registration of keywords, registration of derived keywords, and registration of risk words are performed as needed for maintenance such as addition and deletion, their context is not limited to the above order.

  Step S111: In order to increase the effectiveness of the keyword search, for example, for the search keyword “ABC Shoji Co., Ltd.”, “ABC Shoji”, “ABC Co., Ltd.”, “○ BC Shoji Co., Ltd.”, “AΔC Like derivative expressions such as “Shosho” and “Axx Shoji Co., Ltd.”, fluctuations such as omission of characters such as “Co., Ltd.”, hidden characters, padding, conversion mistakes, and typographical errors are also generated as derived keywords.

Step S115: Using a search program such as a search engine or a search API provided on the website of the information providing medium, one or a plurality of registered keywords and their derived keywords are used as search keywords on the Internet. Search a large number of web pages.
Step S121: Also, the websites of the respective information providing media are crawled by using search engines provided independently on major portal sites, provider sites, news providing sites, blog sites, bulletin board sites, etc. on the Internet.

Step S125, 131: The website of the information providing medium is crawled and the link URL posted on the website is accumulated. The stored link URL is crawled.
Step S135: The data obtained by the search in steps S115, S121, and S131, the data obtained by crawling the medium site, and the data obtained by crawling the link URL are stored.
Step S141: Analyzing accumulated data. For example, for each registered keyword, the location of the published web page, the risk word displayed nearby, the published page, the domain of the published page, the medium of the published page 9 and analyzing the context of the posted content, etc., calculating the degree of relevance with each risk word found for each search keyword, and calculating the risk level with reference to the data table 110 illustrated in FIG. Do scoring.

Step S145: The analysis result of step S141 is stored in the storage device for each search keyword for which a search request has been received.
Steps S151 and S155: Compared with the previous analysis result, it is determined whether or not there has been a change in priority. For example, many high-risk web pages have been released and have been damaged by reputation. If so, send an alert. This alert is transmitted by sending an email, sending a push notification with a smartphone application, or displaying the alert on a personal computer of a contract company. If there is no particularly noticeable change in the determination in step S151, the current reputation damage risk report creation process is stopped without sending an alert.

  In the case of this embodiment, since the degree of risk word relevance is also taken into account, the following reputational damage risk report can be created. Now, the search keyword is “ABC Shoji Co., Ltd.”, and the portal site abcd1. co. It is assumed that a search is performed by a search engine on jp and the following search result is obtained.

  About the search keyword “ABC Shoji Co., Ltd.”, abcd1. co. When “ABC Shoji Co., Ltd.” was entered in the search window using the search engine on jp, “ABC Shoji Co., Ltd. Black Company” was displayed in the third place in the A column displaying the input candidates.

  When a search is performed using “ABC Shoji Co., Ltd.” as a keyword, a large number of web pages including “ABC Shoji Co., Ltd.” are extracted, and the search results are divided into two pages. Web pages were also extracted at the same time, and the URL links of those web pages were displayed along with a few lines of excerpts.

  Furthermore, it is assumed that the notation “ABC Shoji Co., Ltd. Black Company” is found fifth in the B column displaying the related search word on the first page of the search result. In addition, for each web page of the search result, the web page including the risk word “black company” is the first page of the search result, the 2nd and 10th positions in the C column displaying the organic search result, and the 2nd page. I was found in fifth place.

  If the above is the search result, the reputation damage risk level is calculated as follows. Reputation damage risk level = (medium score × weight + page score × weight + placement score × weight) × risk word relevance level.

  Applying this formula to the above example, the media score is the famous portal site abcd1. co. Since it is jp, the media score of the posting site domain is 100 × weight 0.3 = 30. For the first page of the search results, the posted page score is 100 × weight 0.5 = 50, and the second page of the C column is the posted location score 85 × weight 0.8 = 68, C For the web page in the 10th place in the column, the posted part score is 5 × weight 0.8 = 4.

  Further, for the web page in the fifth column of column C in the second page of search results, the media score of the posted site domain is 100 × weight 0.3 = 30, and the page score of search results is 20 × weight 0.5. = 10, placement location score 60 × weight 0.8 = 48. And since it is the 3rd place in the A column, the media score of the posting site domain 100 × weight 0.3 = 30, the posting page of the search result is the first page, the posting page score 100 × weight 0.5 = 50, posting Location score 80 × weight 0.8 = 64.

  In addition, the risk word “black company” has a high risk of negative impact on both youth and investors. In this example, it is assumed that a risk word “ABC Shoji Co., Ltd.” and a “black company” appear nearby and the risk word relevance is determined to be 0.9.

  From such a search result, when the reputation monitoring requesting company is “ABC Shoji Co., Ltd.”, a reputation damage risk report 120 as shown in FIG. 10 is created. However, the table is arranged in descending order of the reputation damage risk level.

Note that the report 120 as shown in FIG. 10 can be displayed on the computer system 101 on the client side by access from the client company. FIG. 8 shows a processing flow in that case. That is, when a client company having an account accesses the system 101 for viewing, authentication processing is performed (step S201), and the keywords requested by the client company, client company information, and importance are displayed in a list. (Step S205)
If there is a request for detailed information from the requesting company, the data storage unit 102 is accessed through the reputation damage risk report creation unit 116 to create a detailed reputation damage risk report 120 for the requesting company (step S211). Then, it is distributed to the client company's computer system via the Internet 100 (step S215).

  As described above, according to the reputation damage risk report creation system, method and program of the present embodiment, detailed reputation damage risk that automatically monitors the Internet and quantifies the degree of reputation damage risk received by the client company. You can create a report. In addition, changes in the reputational damage risk level can be confirmed through periodic searches.

1,101 Computer system 2 Storage device 10, 110 Data table 11, 111 Impact index value storage unit 12, 112 Risk word storage unit 13, 113 Keyword registration unit 14, 114 Risk expression search unit 15, 115 Reputation damage risk level Calculation unit 16,116 Reputation damage risk report creation unit 20,120 Reputation damage risk report 100 Internet

Claims (15)

The influence index value of each of the plurality of information providing media providing information on the computer screen and the influence index value of the place where the search keyword is placed in each of the plurality of information providing media are held as the influence index values. An influence index value storage unit;
A risk word storage unit for holding a plurality of risk words;
A search in which information on the Internet is searched by inputting a specific keyword and a search program is entered, and one of a plurality of risk words held by the risk word storage unit is input in the place where the extracted search results are posted A risk-containing expression search unit for identifying a place where a risk-containing expression appearing with a keyword is found;
For the place of the risk expression specified by the risk expression search unit, the data of the influence index value storage unit is referred to, and the influence index value of the information providing medium providing the information and the search result Reputation damage risk level calculation unit that calculates the impact index value of the posting location and calculates the total of these impact index values as the reputation damage risk level,
The input search keyword and the extracted risk word, the reputation damage risk degree calculated by the reputation damage risk degree calculation unit, the information related to the extracted information providing medium, and the publication location of the risk expression A reputation damage risk report creation system comprising a reputation damage risk report creation section that creates and outputs a reputation damage risk report including information.   The reputation damage risk report creation system according to claim 1, wherein the information providing medium is a portal site, and the search program is a search engine provided on the portal site.   When the reputation damage risk level calculation unit finds the risk-containing expressions for the input keyword in a plurality of locations, the reputation damage risk report creation unit calculates the risk-containing expressions in the plurality of locations in descending order of the reputation damage risk level. The reputation damage risk report creation system according to claim 1 or 2, wherein a reputation damage risk report in a list form is created and output. The influence index value storage unit stores a score and a weight for each of the plurality of information providing media as an influence index value, and a score for each page on which the relevant risk expression is posted for a place where the search result is posted And the weight, and for the display in the input candidate display frame for the input search keyword, the score and the weight are stored for each display line in the input candidate display frame, and the display for the related search word column is displayed. Stores the score and weight for each rank, stores the score and weight for each special module for display on any of the multiple special modules,
The reputation damage risk degree calculation unit calculates a sum of products of the respective scores and weights depending on which place in which information providing medium is the place of inclusion of the risk expression, and the degree of reputation damage risk The reputation damage risk report creation system according to any one of claims 1 to 3, wherein the system is calculated. The risk-containing expression search unit also extracts an expression that approximates one of a plurality of risk words stored in the risk word storage unit, and calculates a coefficient according to the degree of association with the registered risk word. Set,
The reputation damage risk report creation system according to claim 4, wherein the reputation damage risk degree calculation unit multiplies the total by the coefficient to obtain a reputation damage risk degree. A search step of entering specific keywords and searching information on the Internet by a search program;
In the information providing medium that provides the search result extracted by the search step and the posting location, any of the plurality of risk words held by the risk word storage unit appears together with the input search keyword. A risk-containing expression search step for identifying a place where the risk expression is found;
For the posting position of the risk expression specified in the risk expression search step, the influence index value of each of the plurality of information providing media held in the influence index value storage unit and the search keyword in each of the plurality of information providing media By referring to the data of the impact index values of the posting locations, the impact index values of the information providing medium and the impact index values of the posting locations of the search results are determined, and the total of these impact index values is evaluated for damage. Reputation damage risk level calculation step to obtain as a risk level,
The input search keyword and the extracted risk word, the reputation damage risk level obtained in the reputation damage risk level calculation step, the information related to the extracted information providing medium, and the location where the risk expression is posted Reputation damage risk report creation step of creating and outputting a reputation damage risk report including information,
Reputation damage risk report creation program characterized by being executed by a computer.   7. The reputation damage risk report creation program according to claim 6, wherein the information providing medium is a portal site, and the search program is a search engine provided on the portal site.   When the risk damage expression for the input keyword is found in a plurality of places in the reputation damage risk degree calculation step, the reputation damage risk report creation step arranges the risk inclusion expressions in the plurality of places in descending order of the reputation damage risk degree. 8. The reputation damage risk report creation program according to claim 6, wherein a reputation damage risk report in a list format is created and output. The influence index value storage unit stores, as an influence index value, scores and weights for each of the plurality of information providing media, and the search results are posted for each page on which the relevant risk expression is posted. The score and the weight are stored, the display and the weight for each display line in the input candidate display frame are stored for the display in the input candidate display frame for the input search keyword, and the display of the related search word column is Store the score and weight for each display order, store the score and weight for each of the plurality of special modules for display on any of the plurality of special modules,
In the reputation damage risk level calculation step, the sum of the product of each score and weight is calculated according to which posting location in which information providing medium is the posting location of the risk expression, and the reputation damage risk level The reputation damage risk report creation program according to any one of claims 6 to 8, characterized by: In the risk-containing expression search step, an expression that approximates any of a plurality of risk words stored in the risk word storage unit is also extracted, and a coefficient is determined according to the degree of association with the registered risk word. Set,
10. The reputation damage risk report creation program according to claim 9, wherein in the reputation damage risk degree calculation step, the total is multiplied by the coefficient to obtain a reputation damage risk degree. Using a computer
A search step of entering specific keywords and searching information on the Internet by a search program;
In the information providing medium that provides the search result extracted by the search step and the posting location, any of the plurality of risk words held by the risk word storage unit appears together with the input search keyword. A risk-containing expression search step for identifying a place where the risk expression is found;
For the posting position of the risk expression specified in the risk expression search step, the influence index value of each of the plurality of information providing media held in the influence index value storage unit and the search keyword in each of the plurality of information providing media By referring to the data of the impact index values of the posting locations, the impact index values of the information providing medium and the impact index values of the posting locations of the search results are determined, and the total of these impact index values is evaluated for damage. Reputation damage risk level calculation step to obtain as a risk level,
The input search keyword and the extracted risk word, the reputation damage risk level obtained in the reputation damage risk level calculation step, the information related to the extracted information providing medium, and the location where the risk expression is posted Reputation damage risk report creation step of creating and outputting a reputation damage risk report including information,
Reputation damage risk report creation method, characterized in that it is executed and automatically creates a reputation damage risk report.   12. The reputation damage risk report creation method according to claim 11, wherein the information providing medium is a portal site, and the search program is a search engine provided on the portal site.   When the risk damage expression for the input keyword is found in a plurality of places in the reputation damage risk degree calculation step, the reputation damage risk report creation step arranges the risk inclusion expressions in the plurality of places in descending order of the reputation damage risk degree. 13. The reputation damage risk report creation method according to claim 11 or 12, wherein a reputation damage risk report in a list format is created and output. The influence index value storage unit stores, as an influence index value, scores and weights for each of the plurality of information providing media, and the search results are posted for each page on which the relevant risk expression is posted. The score and the weight are stored, the display and the weight for each display line in the input candidate display frame are stored for the display in the input candidate display frame for the input search keyword, and the display of the related search word column is Store the score and weight for each display order, store the score and weight for each of the plurality of special modules for display on any of the plurality of special modules,
In the reputation damage risk level calculation step, the sum of the product of each score and weight is calculated according to which posting location in which information providing medium is the posting location of the risk expression, and the reputation damage risk level The reputation damage risk report creation method according to any one of claims 11 to 13, characterized by: In the risk-containing expression search step, an expression that approximates any of a plurality of risk words stored in the risk word storage unit is also extracted, and a coefficient is determined according to the degree of association with the registered risk word. Set,
15. The reputation damage risk report creation method according to claim 14, wherein in the reputation damage risk degree calculation step, the total is multiplied by the coefficient to obtain a reputation damage risk degree.

Images