JP2017016485A - Calculation device, calculation method, and calculation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To calculate the credibility of a business operator with high accuracy.SOLUTION: A calculation device according to the present application includes an acquisition unit and a calculation unit. The acquisition unit acquires first data, i.e. information on the financial situation of a business operator, and second data, i.e. information on the business operator and based on the user's behavior on a communication network. The calculation unit calculates an index indicating the credibility of the business operator on the basis of the first data and second data acquired by the acquisition unit.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a calculation device, a calculation method, and a calculation program.

  In general, when a financial institution such as a bank determines a loan allowance frame, business data provided by a data provider specializing in collection and analysis of business (company) information is referred to. Such business operator data is generated based on, for example, business operators' financial statements (for example, financial statements, profit and loss statements, etc.). That is, the financial institution uses the business data provided by the business in order to calculate the credit information for determining whether or not the loan amount to each business is appropriate.

  Here, not only data with long update intervals (static data) such as financial data published in the fiscal year end, but also models with input of short update intervals (dynamic data) such as stock prices are created. And the system which can perform the newest company evaluation in time is proposed (for example, patent document 1). In addition to the rapid spread of the Internet in recent years, in addition to information based on business operators' financial statements as in the past, information users share information about businesses on the Internet, and the information they share is unified. A technique for enhancing the objectivity of business operator data by performing integrated management is known (for example, Patent Document 2).

JP 2000-348015 A JP 2010-26602 A

  However, with the above-described conventional technology, it is not always possible to calculate the creditworthiness with respect to the operator with high accuracy. Specifically, the above-mentioned conventional technology only shares information such as financial information of publicly disclosed business operators, business results of business operators that information users can know, and industry news. Even if they are integrated, it is difficult to appropriately evaluate the creditworthiness of the operator (for example, the growth potential and stability of the operator over the future). In addition, since data such as stock prices is not disclosed to unlisted companies, in some cases, there is a shortage of data for corporate evaluation, making it difficult to appropriately calculate the creditworthiness of the business operator.

  The present application has been made in view of the above, and an object of the present invention is to provide a calculation device, a calculation method, and a calculation program capable of calculating the creditworthiness with respect to an operator with high accuracy.

  The calculation apparatus according to the present application includes first data that is information related to the financial status of the operator, and second data that is information related to the operator and information based on user behavior on the communication network. And an calculating unit that calculates an index value indicating the creditworthiness of the operator based on the first data and the second data acquired by the acquiring unit. Features.

  According to one aspect of the embodiment, there is an effect that the creditworthiness with respect to the operator can be calculated with high accuracy.

FIG. 1 is a diagram illustrating an example of a calculation process according to the embodiment. FIG. 2 is a diagram illustrating a configuration example of the calculation apparatus according to the embodiment. FIG. 3 is a diagram illustrating an example of the business operator data storage unit according to the embodiment. FIG. 4 is a diagram illustrating an example of a search information table according to the embodiment. FIG. 5 is a diagram illustrating an example of the site information table according to the embodiment. FIG. 6 is a diagram illustrating an example of a word information table according to the embodiment. FIG. 7 is a diagram illustrating an example of a product information table according to the embodiment. FIG. 8 is a diagram illustrating an example of a social information table according to the embodiment. FIG. 9 is a diagram illustrating an example of a customer information table according to the embodiment. FIG. 10 is a diagram illustrating an example of a model storage unit according to the embodiment. FIG. 11 is a flowchart illustrating a generation processing procedure performed by the calculation apparatus according to the embodiment. FIG. 12 is a flowchart illustrating a calculation processing procedure performed by the calculation apparatus according to the embodiment. FIG. 13 is a diagram illustrating an example of a reception process according to the modification. FIG. 14 is a hardware configuration diagram illustrating an example of a computer that realizes the function of the calculation device.

  Hereinafter, a mode for carrying out a calculation device, a calculation method, and a calculation program according to the present application (hereinafter referred to as “embodiment”) will be described in detail with reference to the drawings. Note that the calculation device, the calculation method, and the calculation program according to the present application are not limited by this embodiment. In the following embodiments, the same portions are denoted by the same reference numerals, and redundant description is omitted.

[1. Example of calculation process)
First, an example of calculation processing according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of a calculation process according to the embodiment. In FIG. 1, the calculation process by the calculation system 1 according to the embodiment will be described. Specifically, FIG. 1 illustrates an example in which calculation processing using a calculation formula (model) that outputs an index value (score) indicating the creditworthiness of an operator is performed by the calculation device 100 included in the calculation system 1. To do.

  As shown in FIG. 1, the calculation system 1 includes a user terminal 10, a financial institution server 30, a web server 40, an operator data server 50, and a calculation device 100. The calculation device 100 is communicably connected to the user terminal 10, the financial institution server 30, the web server 40, and the business operator data server 50 via a communication network (for example, the Internet) (not shown). In addition, the number of the user terminals 10, the financial institution server 30, the web server 40, and the provider data server 50 included in the calculation system 1 is not limited to the example illustrated in FIG. For example, the calculation system 1 may include a plurality of financial institution servers 30, a plurality of web servers 40, and a plurality of business operator data servers 50.

  The user terminal 10 is an information processing device used by general users. Specifically, the user terminal 10 is used for a user to browse a web page or post an evaluation of product information on a website. The user terminal 10 is, for example, a mobile terminal such as a smartphone, a tablet terminal, a PDA (Personal Digital Assistant) or a wearable terminal, a desktop PC (Personal Computer), a notebook PC, or the like. Note that the general user is a user who does not execute an action having a specific intention for the calculation processing according to the embodiment. That is, in the embodiment, a person who professionally acquires and analyzes company information and a person who provides financing to a business may be excluded from general users. Moreover, although the business operators themselves can be excluded from the general users, the business owners and individual officers of the business operators may be included in the general users.

  The financial institution server 30 is a server device used by a financial institution. Specifically, the financial institution server 30 accepts a loan offer from a business operator or notifies the business operator of the result of the offer. In addition, the financial institution server 30 uses the calculation device 100 in order to evaluate the creditworthiness of the business in the financing to the business.

  The web server 40 is a server device that provides various web pages when accessed from the user terminal 10. The web server 40 provides various web pages related to, for example, a news site, a weather forecast site, a shopping site, a finance (stock price) site, a route search site, a map providing site, a travel site, a restaurant introduction site, and a web blog.

  The web server 40 stores user behavior on the network. The user behavior information is stored as user behavior data 42 in the web server 40 or a predetermined storage device. Here, the user behavior on the network means information transmitted from the user terminal 10 in accordance with a user operation when using services provided from various websites. For example, the user behavior on the network is input of a search query on a search site, purchase behavior on a shopping site, posting of a review from a user on a product evaluation site, and the like. Further, the user behavior includes, for example, message exchange on an SNS (Social Networking Service) site, follow-up behavior with respect to others, and the like.

  The provider data server 50 is a server device managed by the provider data provider. A provider data provider, for example, specializes in collecting information related to the financial status of the provider (hereinafter referred to as “financial information”) and calculating the creditworthiness of the provider based on the collected information. It is a company. The provider data provider performs a business of selling the credit information of the provider such as the calculated credit quality to a financial institution or the like. The business operator data server 50 stores, in the business enterprise data 52, financial information of the business operator and information related to artificial evaluation of the business operator (e.g., evaluation of the personality and leadership of the manager). The business data 52 stores various information regarding the business such as financial information of the business that has been disclosed to the public, capital relationships, and stock price trends. Further, the business operator data 52 includes not only publicly available information but also information that is independently researched and acquired by a researcher who belongs to the business operator data provider, for example.

  The calculation device 100 illustrated in FIG. 1 acquires information on a business operator via a communication network, and an index value indicating the creditworthiness of the business operator based on the acquired information (hereinafter sometimes referred to as “score”). Is a server device that generates a model for outputting. Specifically, the calculation device 100 includes information based on user behavior on the Internet (for example, product evaluation sent by the user about products provided by each business operator, , Online reviews, search information on search sites, etc.). And the calculation apparatus 100 produces | generates the model which outputs an operator's score based on financial information and the information based on the user action on the internet. And the calculation apparatus 100 calculates the score of the provider which becomes a process target using the produced | generated model.

  The calculation device 100 acquires information used for calculation processing via each device included in the calculation system 1. Hereinafter, the calculation process performed by the calculation apparatus 100 will be described along the flow with reference to FIG. 1.

  First, the calculation device 100 acquires data related to the business operator included in the business operator data 52 from the business operator data server 50 (step S11). For example, the calculation device 100 acquires the financial information of the business operator as data related to the business operator. Specifically, the calculation apparatus 100 acquires various types of information such as business registration information, asset information, stock price information, sales, profits, and the number of employees.

  Subsequently, the calculation device 100 acquires information about the user included in the user behavior data 42 from the web server 40 (step S12). For example, the calculation device 100 acquires, as data related to the user, information based on user behavior on the Internet, and information that can be acquired via the Internet, among information related to the business operator. Specifically, the calculation device 100 can determine the number of times that the user terminal 10 has performed a search action using the name of a business operator or a product name provided by the business operator as a search query, the number of website visits provided by the business operator, , Get information on reviews posted on products provided by business operators.

  And the calculation apparatus 100 produces | generates the model for calculating the index value which shows a provider's creditworthiness based on the acquired information (step S13). Although details will be described later, the calculation device 100 uses an existing model (hereinafter referred to as an “existing model”) that evaluates the operator using the financial information of the operator, An evaluation value (hereinafter referred to as “financial score”) is calculated. And the calculation apparatus 100 produces | generates the model for calculating a provider's score by learning the correlation with the information acquired from the provider data server 50, the information acquired from the web server 40, and a financial score. To do. The generated model has, for example, information acquired from the provider data server 50 and information acquired from the web server 40 as inputs, and outputs the operator's score (or a predetermined numerical value for calculating the score). . That is, the calculation device 100 inputs information obtained from the business data server 50 and information obtained from the web server 40 to the generated model, which is information on the business subject to be processed. The score of the operator can be calculated.

  Here, the calculation apparatus 100 receives an inquiry about credit information of a predetermined business from the financial institution server 30 (step S14). For example, when the financial institution server 30 receives a loan application from a predetermined business operator, the financial institution server 30 inquires the computing device 100 about the credit information in order to obtain the credit information of the business operator. In other words, the financial institution server 30 transmits information specifying a predetermined business operator to be processed to the calculation device 100 and requests to calculate the score of the business operator. In this case, the calculation apparatus 100 acquires information on a predetermined provider from the provider data server 50 and the web server 40 via the communication network. And the calculation apparatus 100 calculates the score of a predetermined | prescribed provider by inputting the acquired information into a model (step S15).

  Then, the calculation device 100 transmits the calculated score to the financial institution server 30 (step S16). The calculation apparatus 100 can provide the credit information of the operator to the financial institution server 30 through the above-described series of flows.

  As described above, the calculation apparatus 100 according to the embodiment acquires information related to the financial status of the business operator and information related to the business operator and information based on user behavior on the communication network. And the calculation apparatus 100 calculates the index value which shows the creditworthiness of the said provider based on the acquired information.

  That is, the calculation apparatus 100 according to the embodiment is an index that indicates the usage of a business by taking into account not only the financial information of a company generally used for measuring the credit of the business but also information regarding user behavior on the Internet. Calculate the value. In addition, the calculation device 100 obtains information related to the behavior of general users on the Internet, and evaluates the creditworthiness of the operator using information composed of a huge number of samples. For this reason, since the calculation apparatus 100 can determine the creditworthiness of the business operator based on not only the financial information that is disclosed but also the behavior information of many general users, it can calculate the creditworthiness with high accuracy.

  In general, it is difficult to use the financial information of a business operator as the latest data. This is because the financial information of a business operator is relatively updated once every quarter or half year except for information such as stock prices. That is, in the existing model that calculates the credit quality based on the financial information of the operator, the freshness of the data for calculating the index value is inferior, so it is difficult to calculate the operator's credit accuracy with high accuracy in real time. On the other hand, according to the calculation device 100, since information related to user behavior on the Internet is used as data, it is possible to calculate the creditworthiness of the business operator with high accuracy in real time.

  Even if it is a specialized company such as a provider of provider data, it is difficult to completely obtain the financial information of the provider, except for large companies listed and some excellent small and medium-sized companies. For this reason, the enterprises that can calculate the creditworthiness of the enterprise using the existing model are limited to some famous enterprises. On the other hand, according to the calculation device 100, since information about user behavior on the Internet can be used as data, a score can be calculated using a relatively large number of parameters. The score can be calculated with high accuracy for a person.

  Note that the score calculated by the calculation device 100 is not necessarily used for the purpose of financial support or investment. For example, various sales activities performed by the operator itself (the creditworthiness of an advertiser who distributes advertisements, shopping, etc.) It may be used for the creditworthiness of the site as a member store).

[2. Configuration of calculation device]
Next, the configuration of the calculation apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram illustrating a configuration example of the calculation apparatus 100 according to the embodiment. As illustrated in FIG. 2, the calculation device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The calculation device 100 includes an input unit (for example, a keyboard and a mouse) that receives various operations from an administrator who uses the calculation device 100, and a display unit (for example, a liquid crystal display) that displays various types of information. You may have.

(About the communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card). The communication unit 110 is connected to a communication network by wire or wireless, and transmits and receives information to and from the user terminal 10 and the like via the communication network.

(About the storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 120 according to the embodiment includes a provider data storage unit 121, a network information storage unit 122, and a model storage unit 129. Hereinafter, each storage unit will be described in order.

(About the company data storage unit 121)
The business operator data storage unit 121 stores information related to business operators. For example, information stored in the provider data storage unit 121 is provided from the provider data server 50. Here, FIG. 3 shows an example of the business operator data storage unit 121 according to the embodiment. As illustrated in FIG. 3, the business operator data storage unit 121 includes items such as “business operator ID”, “information update date”, “industry”, “financial information”, and “financial score”. The item “financial information” includes small items such as “total assets”, “total liabilities”, “sales”, “operating income”, and “market capitalization”.

  “Business operator ID” indicates identification information for identifying the business operator. “Information update date” indicates the date on which information on the business operator was updated. For example, the information update date indicates the date on which the operator data server 50 has updated the financial information related to the operator. “Industry” indicates the industry or business type to which the business operator belongs.

  “Financial information” indicates information related to the financial status of the operator. For example, the financial information is updated on the basis of data or the like published from the business operator every quarter. In the example of FIG. 3, “total assets”, “total liabilities”, “sales”, “operating income”, and “market capitalization” are illustrated as small items of financial information. The financial information stored in the unit 121 is not limited to the example shown in FIG. For example, the financial information may include information updated every day, such as “stock price”. The financial information may include various types of information provided from the provider data provider. For example, the financial information may include various types of information such as business relationships of business operators, capital adequacy ratio, collateral capacity, and settlement status.

  In the example shown in FIG. 3, items such as “total assets” are represented by the concept “aaa”, but actually, each item has specific information such as the amount corresponding to the total assets. Is memorized.

  “Financial score” indicates a financial score calculated based on financial information. The financial score is calculated using, for example, a predetermined existing model. In the example of FIG. 3, the financial score is represented by a numerical value from 0 to 100. For example, information indicating that the probability of bankruptcy is lower as the financial score is closer to 100 is shown. Note that the financial score may be calculated by the generation unit 132 in the process of model generation to be described later, or may be calculated in advance by the provider data provider.

  In other words, in FIG. 3, as an example of stored information, the company identified by the company ID “A01” is notified by the industry as “Manufactured (Electric machinery) as information updated on“ July 1, 2015 ”. As financial information, for example, “total assets” is “aaa”, “total liabilities” is “bbb”, “sales” is “ccc”, and “operating income” is “ddd”. ”,“ Market capitalization ”is“ eeee ”, and the financial score calculated based on the financial information is“ 80 ”.

  In the following, the identification information stored in the “operator ID” shown in FIG. 3 may be used as a reference symbol. For example, in the following, the business operator identified by the business operator ID “A01” may be referred to as “business operator A01”.

(About the network information storage unit 122)
The network information storage unit 122 stores user information acquired via a communication network. Specifically, the network information storage unit 122 stores information related to a business operator and based on user behavior on the communication network. As shown in FIG. 2, the network information storage unit 122 includes a search information table 123, a site information table 124, a word information table 125, a product information table 126, a social information table 127, and a customer information table 128. Have. In the following, information that is stored in each data table and is related to a business operator and based on user behavior on a communication network may be collectively referred to as “network information”. Hereinafter, each data table will be described in order.

(Regarding the search information table 123)
The search information table 123 stores information related to user search behavior on the Internet. Here, FIG. 4 shows an example of the search information table 123 according to the embodiment. As shown in FIG. 4, the search information table 123 includes items such as “operator ID”, “total period”, “number of searches”, “degree of increase”, “search ranking”, and “target word”.

  “Business operator ID” indicates identification information for identifying the business operator. “Aggregation period” indicates a period in which actions performed by the user terminal 10 are aggregated. In the example of FIG. 4, the aggregation period is set to one week, but the aggregation period may be a different period. For example, by setting the total period to one month, the calculation device 100 can easily grasp the search tendency over a longer period.

  “Number of searches” indicates, for example, the number of times that a searcher has been searched by a search engine on a predetermined search site. The search query counted as the number of searches is not limited to the name of the business operator itself, and may include the name of a product provided by the business operator, the name of the business operator of the business operator, and the like.

  The “degree of increase” indicates an increase / decrease value of the number of searches with respect to the number of searches in the immediately preceding aggregation period. “Search ranking” indicates a ranking of the number of searches in a predetermined search site. Although illustration is omitted, the search ranking item may store not only the ranking based on the number of searches but also the ranking based on the degree of increase.

  The “target word” indicates a word that is counted as the number of searches related to the operator when the described word is transmitted as a search query. For example, when the name of a product provided by a business is higher than the name of the business, the user may perform a search action using the product name. In this case, even if a search is performed using the product name as a search query, if the product name is set as the target word, the number of searches for the business provider that provides the product is counted. The target word may be set artificially by an application of the administrator of the calculation apparatus 100 or a business operator, or may be automatically set by analyzing a link of a website based on a search result, for example. Good. Specifically, when searching for “product B01”, if a large number of websites related to “company name A01” are displayed as search results, “product B01” is displayed together with the company name A01. Is automatically set to the target word.

  That is, in FIG. 4, as an example of data, the business operator identified by the business operator ID “A01” is “30000 times” during the aggregation period “June 15th 2015 to June 21st 2015”. The degree of increase from the number of previous searches is “1000 times”, and the search ranking is “12000”. In addition, it is shown that “enterprise name A01”, “product B01”, “manager C01”, and the like are set in the enterprise A01 as target words in the search.

  As another example, the operator A11 searches for “200000 times” during the aggregation period “June 30, 2015 to July 6, 2015”, and the number of searches for “195500 times” It shows that it is rising. This is presumed that an event in which the name recognition of the business operator A11 rapidly increased occurred. In such a case, the search information table 123 may store, for example, a mark for recognizing the business operator A11 as a notable company according to the rapid rise. Such a mark may be used, for example, for generating a model to be described later.

(About the site information table 124)
The site information table 124 stores information related to websites operated or managed by the business operator. Here, FIG. 5 shows an example of the site information table 124 according to the embodiment. As shown in FIG. 5, the site information table 124 includes items such as “operator ID”, “total period”, “PV”, “UU”, and “CVR”.

  The “company ID” and “total period” correspond to the same items stored in the search information table 123. “PV” indicates a page view on the website, that is, the number of browsing.

  “UU” indicates the number of unique users. The unique user indicates the number of visitors to the website. Even if the same user visits the same website many times, the UU number is counted as “1”.

  “CVR” indicates a conversion rate. CVR indicates the percentage of conversions to the number of browsing websites. Conversion means the final result you can get on your website. For example, the conversion is purchase of a product on an online shopping site, member registration on an information providing site or a community site, request for materials, and the like. The CVR may be a conversion rate for the number of browsing, or a conversion rate for a unique user may be adopted.

  That is, in FIG. 5, as an example of data, the website provided by the operator A01 is viewed “11000 times” during the aggregation period “June 15, 2015 to June 21, 2015”. The number of UUs viewed is “3000”, indicating that conversion has been performed by “1%” of the number of browsing.

  Note that the tabulation period in FIG. 5 is an example, and PV and the like may be tabulated not on a weekly basis but on a daily basis or on a monthly basis. In addition, for each item such as PV, not only the numerical value as an absolute number but also the number of fluctuations from the immediately preceding aggregation period may be aggregated.

(Regarding the word information table 125)
The word information table 125 stores information about words transmitted from users on the Internet. Here, FIG. 6 shows an example of the word information table 125 according to the embodiment. As shown in FIG. 6, the word information table 125 includes items such as “operator ID”, “total period”, “keyword appearance number”, “co-occurrence word”, and “co-occurrence word appearance number”.

  The “company ID” and “total period” correspond to the same items stored in the search information table 123.

  “Keyword Appearance Number” indicates the number of words (keywords) related to business operators that have appeared on the Internet. The word relating to the business operator corresponds to, for example, the “target word” stored in the search information table 123. The number of appearances on the Internet indicates, for example, the number of times a keyword has been posted from the user terminal 10 to the web server 40. In this case, the website provided by the web server 40 includes not only a search site but also various websites such as a short text posting site and a blog providing site.

  The “co-occurrence word” indicates a word used together with the keyword when the keyword appears. For example, a co-occurrence word corresponds to a word posted as a search query together with a keyword on a search site. The co-occurrence word also corresponds to a word that appears together with a keyword in one short sentence on a short sentence posting site. The “co-occurrence word appearance count” indicates the number of times that a predetermined word appears on the Internet together with the keyword.

  That is, in FIG. 6, as an example of data, a keyword related to the business operator identified by the business operator ID “A01” is “55000” during the aggregation period “June 15, 2015 to June 21, 2015”. The word that appears “times” and co-occurs with the keyword indicates “good”, “financial settlement”, “product B01”, and the like. Further, for example, the number of occurrences of “good” co-occurring with the keyword is “3000”.

  Although not shown in FIG. 6, the word information table 125 may be provided with items such as “degree of increase” indicating the transition of the number of keyword appearances, similarly to the search information table 123. In other words, the word information table 125 may store information related to changes in the number of keywords appearing on the Internet.

  Further, the word information table 125 may count the appearance of words for each predetermined document (document) and store characteristic information included in the document. For example, the word information table 125 stores, for each short sentence posted on the short sentence posting site, a word extracted as a word characterizing the short sentence as a keyword and the number of appearances. For example, a keyword characterizing a short sentence is obtained by using an algorithm that scores each word appearing in the short sentence, such as tf-idf (Term Frequency-Inverse Document Frequency). In this case, it is assumed that the number of occurrences of words on the Internet is not counted at random, but the keywords related to the business are counted when a matter related to the business is described in a short sentence. . According to this method, the number of occurrences of keywords related to a business operator is counted only when a matter related to the business operator is posted on the Internet with a certain intention of the user. Data. The word information table 125 may store the weight of the co-occurrence word that co-occurs with the keyword in a predetermined document. The weight of the co-occurrence word corresponds to, for example, a numerical value scored by tf-idf or the like, a predetermined index value based on the appearance rate of the co-occurrence word, and the like.

  Further, the word information table 125 may store the characteristics of co-occurring words. The characteristics of co-occurring words are, for example, “good” or “surplus”, or the nature of giving a positive impression by co-occurring with a business operator, or “negative” or “red” This refers to the nature of co-occurring words, such as nature. For example, in a score calculation process to be described later, an operator having a high ratio of co-occurring with a word having a positive characteristic tends to be calculated with a high score indicating trustworthiness.

(About the product information table 126)
The product information table 126 stores information related to products provided by the business operator. Here, FIG. 7 shows an example of the product information table 126 according to the embodiment. As illustrated in FIG. 7, the product information table 126 includes items such as “enterprise ID”, “product”, “user evaluation”, “number of reviews”, and “store ranking”.

  “Business operator ID” corresponds to the same item stored in the search information table 123. “Product” indicates the name of the product provided by the operator.

  “User evaluation” indicates an evaluation value indicated by a general user in a product evaluation site on the Internet. The product evaluation site is, for example, a community site that receives word-of-mouth information such as product reviews and evaluations from general users. In addition, when a product provided by a business operator is an application for a terminal, a site (referred to as an application store) that provides an application download service may also serve as a product evaluation site. Here, user evaluation shall be shown by the average value of the numerical value from "0" transmitted from a user to "5".

  “Number of reviews” indicates the number of reviews posted by the user at the product evaluation site on the Internet. “Store ranking” indicates a ranking among similar products handled in the product evaluation site. The store ranking may be ranked according to the numerical value of the user evaluation, or may be ranked according to the number of products sold. As described above, when the product evaluation site is an app store, the store ranking may be the number of downloads of the app.

  That is, in FIG. 7, as an example of data, the product “B01” provided by the operator A01 has a user evaluation of “4 points”, the number of reviews posted is “4500”, and the store ranking is “4”. "10th place".

(About social information table 127)
The social information table 127 stores an index value that can be used to evaluate the business's social reputation and personality. Specifically, the social information table 127 stores information acquired from an SNS site used by business owners and officers. Here, FIG. 8 shows an example of the social information table 127 according to the embodiment. As illustrated in FIG. 8, the social information table 127 includes items such as “enterprise ID”, “investigation subject”, and “SNS connection number”.

  “Business operator ID” corresponds to the same item stored in the search information table 123. “Survey object person” indicates an individual name to be subjected to social analysis. For example, the survey target is a manager of the business operator, an executive such as a president or a director.

  The “SNS connection number” indicates a numerical value connected to another person on the SNS when the survey target person uses the SNS. For example, the number of followers on the SNS corresponds. Note that the number of SNS connections may not include the number of general users, but may count only the number of connections between managers and officers of different operators. Thereby, the number of SNS connections can be a more reliable numerical value as an index value indicating the human pulsation force in the survey subject.

  That is, in FIG. 8, as an example of the data, the survey target person in the business operator A01 is “C01” or “C02”, the number of connections in the SNS used by “C01” is “120”, and “C02” This indicates that the number of connections in the SNS used by is “50”.

(About customer information table 128)
The customer information table 128 stores information related to the business customers. Here, FIG. 9 shows an example of the customer information table 128 according to the embodiment. As shown in FIG. 9, the customer information table 128 includes items such as “business ID”, “number of product users”, “continuous usage rate”, and “customer unit price”.

  “Business operator ID” corresponds to the same item stored in the search information table 123. “Number of product users” indicates the number of customers who use the products provided by the business. For example, when the product provided by the business is an application, the total number of downloads of the application provided by the business corresponds.

  “Continuous usage rate” indicates the continuous usage rate of the customer with respect to the operator. For example, when a business operator operates a shopping site on the Internet, the ratio of the number of users who regularly use the site corresponds to the total number of users browsing the site. Further, the continuous usage rate may be the number of user terminals 10 whose use is continuously confirmed with respect to the total number of downloads when the provider provides an application. In this case, the continuous usage rate is stored as an operation rate of the application (for example, a value obtained by dividing the number of usage users per predetermined period by the number of download users).

  “Customer unit price” indicates a unit price per customer. For example, when a business provider provides a shopping site, the customer unit price corresponds to the purchase amount per user in a predetermined period. The unit price of the customer can also be calculated based on the download sales amount of the application, the cost for continuous use, and the like when the provider provides the application.

  That is, in FIG. 9, as an example of data, the product user provided by the operator A01 is “300000”, the continuous usage rate is “0.25”, and the customer unit price is “8000 yen”. Is shown.

(About the model storage unit 129)
The model storage unit 129 stores information regarding the model generated by the calculation device 100. Here, FIG. 10 illustrates an example of the model storage unit 129 according to the embodiment. As illustrated in FIG. 10, the model storage unit 129 includes items such as “model ID”, “information update date”, and “industry”.

  “Model ID” indicates identification information for identifying a model. “Information update date” indicates the date on which the model was updated. “Industry” indicates the industry to which the business operator whose score is calculated belongs. This means that models are generated for each industry of operators. That is, the model is generated by using common business data of a predetermined industry. This is because it is easier to find the commonality and similarity of numerical values in the items to be compared when using business data of the same industry when generating the model.

  That is, in FIG. 10, as an example of data, the model M001 has information updated on “July 13, 2015”, and the industry to which the model M001 is applied is “manufacturing (electrical)”. ing.

(About the control unit 130)
The control unit 130 is configured such that, for example, various programs (corresponding to an example of a search program) stored in a storage device inside the calculation device 100 are stored in a work area by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. This is realized by being executed as The control unit 130 is realized by an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

  As shown in FIG. 2, the control unit 130 according to the embodiment includes an acquisition unit 131, a generation unit 132, a reception unit 133, a calculation unit 134, and a notification unit 135, and information processing described below Realize or execute the functions and operations of The internal configuration of the control unit 130 is not limited to the configuration illustrated in FIG. 2, and may be another configuration as long as the information processing described later is performed. In addition, the connection relationship between the processing units included in the control unit 130 is not limited to the connection relationship illustrated in FIG. 2, and may be another connection relationship.

(About the acquisition unit 131)
The acquisition unit 131 acquires information related to the business operator from information on the communication network. For example, the acquisition unit 131 acquires information related to the financial status of the operator via the operator data server 50. Moreover, the acquisition part 131 acquires the financial score which is an evaluation value calculated based on a provider's financial information, for example by utilizing an existing model. Moreover, the acquisition part 131 is information relevant to a provider, and acquires information based on user behavior on the communication network.

  Specifically, the acquisition unit 131 according to the embodiment specifies a business operator serving as a model generation sample, and searches the Internet for information regarding the specified business operator. And the acquisition part 131 acquires the information transmitted from the user at the time of use of the service which a provider relates among the services provided from various websites as information based on the user action on the Internet from the web server 40. . Note that information based on user behavior on the Internet includes, for example, transmission of a search query performed by a user on a search site, review of a product performed by a user on a product evaluation site, and release of information on a SNS by a user. It means information generated when a user uses a service on various websites or information transmitted from the user terminal 10. The service related to the business is not limited to a service such as a shopping site directly provided by the business. For example, a service provided by a search site capable of searching for the business or evaluation of the business of the business. This includes services provided by evaluation sites that can be used.

  For example, the acquisition unit 131 acquires search information regarding a business operator at a predetermined search site. Specifically, the acquisition unit 131 acquires search information such as how many times the business has been searched by the user, using a search query related to the business as a search target word. Then, the acquisition unit 131 stores the acquired information in the search information table 123.

  In addition, the acquisition unit 131 acquires site information on a website provided by a business operator that is an acquisition target of information. Specifically, the acquisition unit 131 acquires information such as the number of PVs, the number of UUs, and CVR of the website provided by the business operator. Then, the acquisition unit 131 stores the acquired information in the site information table 124.

  In addition, the acquisition unit 131 acquires the number of occurrences of words on a website provided by a business operator that is an acquisition target of information. Specifically, the acquisition unit 131 identifies keywords related to the business operator, and acquires the number of occurrences of the identified keywords, the types of words that co-occur with keywords, and the number of occurrences of words that co-occur with keywords. To do. Then, the acquisition unit 131 stores the acquired information in the word information table 125.

  In addition, the acquisition unit 131 acquires information related to a product provided by a business operator from which information is acquired. Specifically, the acquisition unit 131 acquires information such as user evaluation, the number of reviews, and store ranking on the Internet of products provided by the business operator. Note that the acquisition unit 131 may acquire information (that is, a change rate) regarding a trend of each information such as a change in user evaluation and a change in store ranking. In addition, when the product provided by the provider is a program product such as an application, the acquisition unit 131 displays the number of downloaded applications, the number of users, the average usage time per user, and the applications in a predetermined period. Get the index value such as the utilization rate of. Then, the acquisition unit 131 stores the acquired information in the product information table 126.

  Moreover, the acquisition part 131 acquires the information which can evaluate the social reputation and attractiveness of the provider from which information is acquired. Specifically, the acquisition unit 131 acquires information such as the number of consolidated SNSs of business managers and officers. The acquisition unit 131 may acquire the trends of individual managers on the Internet as index values that can evaluate the social reputation and attractiveness of the business. For example, the acquisition unit 131 acquires information on a person who is personally connected to the manager on the SNS (for example, name recognition, the size of the company to be managed). That is, the acquisition unit 131 acquires information or the like that is assumed to indicate the human network power of the manager. In addition to the above-mentioned number of SNS connections, the acquisition unit 131 determines the number of accesses from general users and the number of followers, etc. You may acquire as information about a name recognition and reputation. Then, the acquisition unit 131 stores the acquired information in the social information table 127.

  Moreover, the acquisition part 131 acquires the information regarding the customer of the provider who becomes the acquisition target of information. Specifically, the acquisition unit 131 acquires information such as the number of users of a product provided by the business operator, the continuous usage rate of the user, and the customer unit price. For example, for a provider that provides an application, the acquisition unit 131 can acquire the number of product users by acquiring the number of downloaded applications from the application store. In addition, for a business operator who operates a shopping site, the acquisition unit 131 acquires information on a continuous usage rate and a customer unit price based on information such as a visit interval of a user's site and a purchase amount. Can do. Then, the acquisition unit 131 stores the acquired information in the customer information table 128.

  Note that the acquisition unit 131 may randomly acquire information on various operators without specifying the operator from which the information is acquired. For example, the acquisition unit 131 uses a program such as a search robot used for a search engine or the like to crawl on the Internet to acquire information about the operator as needed or update the acquired information. Can do.

(About the generator 132)
The generation unit 132 generates a model for calculating the creditworthiness of the business based on various information. Specifically, the generation unit 132 according to the embodiment acquires information related to the financial status of the business operator acquired by the acquisition unit 131 (hereinafter may be referred to as “first data”) and the business. For predicting an index value indicating the creditworthiness of the operator based on information related to the operator and based on information based on user behavior on the communication network (hereinafter sometimes referred to as “second data”) Generate a model of

  When generating the model, the generation unit 132 uses the financial score of the operator calculated in advance based on the existing model. The generation unit 132 may use the financial score provided by the provider data provider acquired by the acquisition unit 131, or calculates a financial score using an existing model as described below. It may be used for processing. The generation unit 132 generates a model based on the correlation between the operator's financial score and the first data and the second data. The model generation process will be described below.

  In addition, the production | generation part 132 can use arbitrary known company evaluation models as an existing model. For example, the Alt value (Edward Altman) Z value and the like are widely known as a company evaluation model for obtaining a score indicating the probability of bankruptcy of an operator based on financial information. The generation unit 132 calculates the financial score of the business operator from the first data acquired by the acquisition unit 131 using such an existing model. In this case, the financial score of the business operator can be arbitrarily changed depending on the existing model used, such as a score indicating the probability of bankruptcy and a predetermined score indicating the financial credibility of the business operator. A formula that simplifies the calculation is represented by the following formula (1), for example.

y ₁ = α ・ (Financial information of the operator) (1)

In the above formula (1), “y ₁ ” represents the financial score of the business operator to be output, “Financial information of the business operator” is the input data, and “α” represents a predetermined weight value. In other words, the generation unit 132 can output and calculate “y ₁ ” that is the financial score of the business operator by inputting the financial information of the business operator. “Financial information of a business operator” can be divided into items obtained by variously combining numerical values such as total assets, total liabilities, sales, and retained earnings. Therefore, the above formula (1) can be expressed by, for example, the following formula (2 ).

y ₁ = α ₁ · x ₁ + α ₂ · x ₂ + α ₃ · x ₃ + ... + α _n · x _n (n is an arbitrary number) (2)

In the above formula (2), “x _n ” represents each item indicating the financial information of the operator, and “α _n ” represents the weight value of each item. As an example, in the above-described model for obtaining the Altman Z value, it is known that “x ₁ ” is (working capital / total assets) and “α ₁ ” is “0.012”. Subsequently, the generation unit 132 establishes an expression such as the following expression (3) as a model to be generated.

Y ₁ = β · (business financial information) + γ · (business network information) (3)

In the above formula (3), “Y ₁ ” is an index value (score) indicating the creditworthiness of the operator in the present application, and “network information of the operator” is information related to the operator stored in the network information storage unit 122. , “Β” and “γ” indicate predetermined weight values. That is, the generation unit 132 generates a model in which “business operator network information” that is information based on user behavior on the communication network is added in addition to “business operator financial information” in the existing model.

For example, when generating a model by regression analysis, the generation unit 132 generates a model by learning using a financial score (“y _n ”) calculated based on the financial information of the business operator as correct answer data. As an example, the generation unit 132 performs learning using the following formulas (4) to (6).

y ₂ = α • (Financial information for a given period of business) (4)

Y ₂ = β · (financial information of the operator for a predetermined period) + γ · (network information of the operator from the predetermined period corresponding to the financial information to the next settlement of accounts) (5)

y ₃ = α • (Financial information at the time of the next fiscal year of the operator) (6)

In this case, the generation unit 132 performs learning for applying “Y ₂ ” in the above equation (5) using “y ₂ ” and “y ₃ ” in the above equations (4) and (6) as correct answer data. As described above, the generation unit 132 performs learning based on the correlation between the financial score obtained from the financial information of the business operator and the first data and the second data. For example, as a result of learning, the generation unit 132 is based on “financial information of the operator for a predetermined period” and “network information of the operator from the predetermined period corresponding to the financial information to the next settlement” in Expression (5). Thus, a model capable of outputting “Y ₂ ” which is the financial score of the business operator at the time of the next fiscal year is generated.

In this case, as learning progresses, it is assumed that in Formula (5), when the first data and the second data are input, a value indicating the trustworthiness of the future operator can be output. The Specifically, the input of “the network information of the operator from the predetermined period corresponding to the financial information to the time of the next settlement” is added to the “financial information of the operator for a predetermined period” in the expression (5). It is assumed that a value corresponding to (approximate) “y ₃ ” that is the value output in (6) is output. This means that the trustworthiness of the operator in the future can be output at the time when the network information input to the model is acquired. In other words, the generation unit 132 can generate a model that accurately outputs the operator's score in real time.

  Note that any information stored in the network information storage unit 122 may be used for the network information of the business operator. That is, the above formula (3) may be expressed as the following formula (7), for example.

Y = β ₁ · x ₁ + β ₂ · x ₂ + β ₃ · x ₃ + ... + β _n · x _n (7)

In the above formula (7), for example, in “x _{1 to} x ₅ ”, an item based on “business operator financial information” corresponds. In addition, for example, in “x _{6 to} x _n ”, “operator network information” corresponds. Actually, the “network information of the business operator” includes variables obtained by quantifying the information acquired by the acquisition unit 131 according to a predetermined standard (for example, the number of searches, the ranking of product evaluation, the number of users of products, etc. , Items represented by numerical information so that they can be compared with each other). For example, “x ₆ ” is an item corresponding to “number of searches”, “x ₇ ” is an item corresponding to “product evaluation”, or “x ₈ ” is “number of product users”. May be an item corresponding to "." Then, as learning progresses, the generation unit 132 optimizes the value of “α _n ” corresponding to each item. That is, for example, the generation unit 132 obtains the value of the weight value “α _n ” such as how much the “number of searches” affects the score in calculating the score indicating the creditworthiness of the operator, Set to model.

  The model described above is an example, and the model generated by the generation unit 132 is not limited to the above example. In other words, the generation unit 132 receives the first data and the second data and outputs a score indicating the creditworthiness of the business operator, so that a model different from the above example can be obtained by combining known methods. You may produce | generate suitably. For example, in the above example, an example in which a model is generated by regression analysis has been shown. However, the generation unit 132 may generate a model by other statistical processing, or a model using a neural network or the like. It may be generated.

  Further, the generation unit 132 may receive an artificial input such as an administrator of the calculation apparatus 100 with respect to the network information used in the model. For example, when the administrator of the calculation apparatus 100 determines that it is appropriate to use “number of searches” indicating the number of searches for a business by a general user as an item used to measure a business operator's score. , Such information is input to the calculation apparatus 100. Then, the generation unit 132 automatically determines an item (for example, an item whose weight value is calculated to be lower than a predetermined threshold) that is determined to be not useful as an item used for measuring a score in the model generation process. Learn to exclude them. Thereby, the generation unit 132 can optimize the generated model.

  In addition, the generation unit 132 may appropriately adjust the period during which information is acquired with respect to selection of network information used for generating a model. For example, the number of searches and the number of PVs on websites fluctuate rapidly due to the influence of positive materials (such as when news is reported in the development of hot products) and negative materials (such as news reporting of fraudulent detection). It is possible. In this case, the generation unit 132 can suppress the influence by using the number of searches acquired over a longer period than usual. Such determination of the period may be manually determined by the administrator of the calculation apparatus 100, or may be automatically determined by analyzing a word of the news article.

  In addition to the number of searches described above, the generation unit 132 can use information stored in the network information storage unit 122 as appropriate to generate a model. For example, the generation unit 132 can use product information analysis, analysis based on the social relationship of the business operator, customer analysis, and the like. Specifically, the generation unit 132 relates to the user evaluation in the product information table 126, the number of reviews, the size of the value stored in the item of the store ranking, the degree of publicity and personality of the business managers and officers, etc. Information is appropriately digitized and used as a variable. In addition, the generation unit 132 appropriately inputs the number of downloaded applications, the number of users, the average usage time per user, the operation rate in a predetermined period, or the fluctuation rate of these values in a predetermined period. A model may be generated that can be used as: The generation unit 132 may perform normalization processing according to the features used for generating these models. The generation unit 132, for example, for the number of searches, the number of PVs and UUs, the number of keyword appearances, the number of co-occurrence keywords, the number of reviews of products and apps, the number of SNS connections, the number of users of products and apps, the unit price of customers, etc. Instead of using the acquired numerical value itself, it is possible to use a numerical value normalized by a value indicating the size of the company such as total assets. It should be noted that the generation unit 132 does not necessarily need to perform normalization regarding features whose numerical values themselves include relative values and ratios, such as ranking related to search, CVR, user evaluation, and continuous usage rate. Further, the generation unit 132 may generate a model based on the importance of words related to the business operator. For example, the generation unit 132 quantifies the importance of a word related to the business operator (for example, the target word shown in FIG. 4), thereby affecting the business operator's score by the word acquired via the network. Can be generated. The generation unit 132, for example, the number of appearances of words related to the business, the number of appearances of words co-occurring with the business, the characteristics of the words co-occurring with the business, the words co-occurring with the business in a predetermined document The importance of the word related to the operator is determined from various information related to the word such as weight or a combination thereof. Specifically, when a word indicating an operator or a product name provided by the operator appears on the network, the generation unit 132 determines the importance of the word (in other words, related to the operator). The predetermined index value indicated by the word to be determined) can be determined high, and a model in which the operator's score is calculated based on the determined importance can be generated. In addition, the generation unit 132 includes a tf-idf of a co-occurrence word in a predetermined document (a predetermined short sentence posted by a user, a predetermined web page, or the like), a predetermined index value based on the appearance rate of the co-occurrence word, or the like. Based on the above, the importance of the word related to the operator may be determined.

  In addition, the generation unit 132 may generate a model using information on business operators belonging to a common industry. For example, as for information such as product reviews, it is assumed that more appropriate information can be obtained by comparing peers than by comparing products of different industries. Therefore, the generation unit 132 classifies the business operators for each industry and generates a model for each industry. In this case, the calculation unit 134, which will be described later, calculates a score using an industry model corresponding to the business target to be processed. FIG. 10 shows an example in which the generation unit 132 generates a model for each industry and stores it in the model storage unit 129.

(About the reception unit 133)
The accepting unit 133 accepts a request regarding the creditworthiness of the business operator. Specifically, the accepting unit 133 according to the embodiment accepts an inquiry regarding the credit information of the operator to be used for setting the financing conditions from the financial institution server 30. In this case, the reception unit 133 may receive information on the business operator along with the offer. For example, the accepting unit 133 accepts the name of the business operator, product information provided by the business operator, the business type of the business operator, information on the manager and officers, and the like. Then, the reception unit 133 sends the received information to the calculation unit 134, which will be described later, thereby calculating the operator's score. Note that the reception unit 133 does not receive product information provided by the business operator or information on managers and officers and the financial institution server 30 does not receive such information from the financial institution server 30, or network information. When such information is not stored in the storage unit 122, information may be newly acquired from an external information processing apparatus (for example, the web server 40). Then, the reception unit 133 sends the received information to the calculation unit 134.

(About the calculation unit 134)
Based on the first data and the second data acquired by the acquisition unit 131, the calculation unit 134 calculates an index value indicating the creditworthiness of the operator. Specifically, the calculation unit 134 according to the embodiment inputs the first data and the second data relating to the business operator received by the reception unit 133 to the model generated by the generation unit 132, so that The score of a certain operator is output. And the calculation part 134 calculates a provider's reliability based on the output score. Note that the calculation unit 134 may use the output score itself as the creditworthiness of the operator.

  For example, when the reception unit 133 receives a request regarding the creditworthiness of the business operator A11, the calculation unit 134 acquires the financial information and network information of the business operator A11. Then, the calculation unit 134 inputs the financial information and network information of the business operator A11 into the model generated by the generation unit 132. And the calculation part 134 acquires the score of provider A11 as an output. In addition, when the model corresponding to the industry to which the business operator A11 belongs exists in the model storage unit 129, the calculation unit 134 may calculate the score by using the model corresponding to the industry preferentially. .

(About the notification unit 135)
The notification unit 135 notifies a response to the received request. Specifically, the notification unit 135 according to the embodiment notifies the credit (credit information) such as the score of the business subject to evaluation in response to the request of the financial institution server 30 received by the reception unit 133.

[3. Processing procedure)
Next, a generation process procedure performed by the calculation apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 11 is a flowchart illustrating a generation processing procedure performed by the calculation apparatus 100 according to the embodiment.

  As illustrated in FIG. 11, the acquisition unit 131 according to the calculation device 100 acquires company data from the company data server 50 (step S101). Moreover, the acquisition part 131 acquires the network information regarding a provider via a communication network (step S102).

  And the production | generation part 132 produces | generates the model for calculating the score which shows the creditworthiness of the said provider based on the information relevant to the financial condition of a provider, and the network information regarding a provider (step S103). The generation unit 132 stores the generated model in the model storage unit 129, and ends the generation process.

  Next, the procedure of the calculation process performed by the calculation apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 12 is a flowchart illustrating a calculation processing procedure performed by the calculation apparatus 100 according to the embodiment.

  As illustrated in FIG. 12, the reception unit 133 according to the calculation device 100 determines whether or not a request for creditworthiness has been received from the financial institution server 30 (step S201). If the request is not received, the reception unit 133 waits until it is received (step S201; No).

  On the other hand, when a request is received (step S201; Yes), the reception unit 133 specifies a business entity to be processed. And the acquisition part 131 acquires the financial information and network information of a provider which become the object of processing (Step S202). Then, the calculation unit 134 inputs financial information and network information about the business acquired by the acquisition unit 131 into the model (step S203).

  And the calculation part 134 outputs a provider's score by inputting the information regarding a provider into a model (step S204). The calculating unit 134 calculates the creditworthiness of the business based on the output score (step S205). And the notification part 135 notifies the calculated result to the financial institution server 30 (step S206), and complete | finishes a calculation process.

[4. (Modification)
The calculation device 100 according to the above-described embodiment may be implemented in various different forms other than the above-described embodiment. Therefore, in the following, another embodiment of the calculation device 100 will be described.

[4-1. (Providing the platform)
In embodiment mentioned above, the calculation apparatus 100 showed the example which produces | generates the model which outputs an operator's score using the financial score calculated based on an operator's financial information. Moreover, the calculation apparatus 100 showed the example which utilizes the provider data server 50 as an acquisition destination of a provider's financial information. However, for a company that is not listed or a business that does not disclose financial information, the calculation device 100 may not be able to acquire financial information of the business subject to be processed.

  Therefore, the calculation device 100 can publish a predetermined platform on which a business operator or a third party can provide information, and accept information related to the business operator. This point will be described with reference to FIG.

  FIG. 13 is a diagram illustrating an example of a reception process according to the modification. As shown in FIG. 13, in the public platform provided by the calculation device 100, information about the business (for example, information stored in the business data storage unit 121) is disclosed to general users and businesses 55.

  At this time, the calculation apparatus 100 may disclose the score calculated for each business operator as information stored in the business operator data storage unit 121. Then, it is assumed that the general user can browse the published information via the user terminal 10. That is, for the general user, the evaluation (credit information) of the business operator updated in real time can be confirmed.

  For the business operator 55, the company's score is disclosed and financial information such as total assets is disclosed. Here, when the business operator 55 is not listed, the company information may be disclosed in a state where the relevant information is missing for a part of the financial information. As described above, when a part of the financial information is lacking, there is a risk that the business operator 55 may have a bad impression from a general user or may be judged as a business enterprise lacking in reliability.

  Therefore, the calculation device 100 receives provision of business operator information from the business operator 55. That is, in the public platform, the calculation device 100 provides an environment in which the business operator 55 can directly or indirectly provide their own information. Thereby, the calculation apparatus 100 can acquire accurate financial information even from a business operator who is not listed and is difficult to acquire financial information. And the calculation apparatus 100 updates a provider's credit information based on the acquired information. Note that the calculation apparatus 100 may appropriately accept information other than financial information.

  As described above, the calculation apparatus 100 acquires the first data and the second data via a predetermined platform on which an operator or a third party can provide information. And the calculation apparatus 100 calculates the score which shows the reliability of the said provider based on the information provided by the provider or the third party.

  Thereby, the calculation apparatus 100 can acquire information on a business operator that is generally difficult to acquire. In addition, when the information update date is disclosed, the general user can grasp that the business operator 55 is appropriately updating the information. Through such behavior, the business operator 55 can increase the trust of general users. In addition, since the operator 55 can increase its own credibility by publishing information on a public platform and can update the score with appropriate information provided by the company, the credibility of the company 55 for general users and financial institutions Can be improved.

[4-2. Data to be input to model)
In the above-described embodiment, the calculation apparatus 100 has input network information related to an operator and calculates an operator's score. Here, the calculation apparatus 100 may associate the acquisition timing of the financial information used for calculating the financial score of the business operator with the acquired network information.

  That is, the calculation device 100 generates a model based on the first data when the financial score is calculated and the second data aggregated at the time corresponding to the first data. Also good.

  Specifically, the calculation apparatus 100 responds when the financial score used to generate the model is calculated using financial information collected from “January 2015” to “March 2015”. The model is calculated using the network information acquired from “January 2015” to “March 2015”, which is the timing of the operation. Thereby, the calculation apparatus 100 can generate a model that appropriately reflects the correlation between the network information acquired from the general user and the financial information. Note that the calculation device 100 does not necessarily need to be the same period as the time corresponding to the financial information. For example, the calculation apparatus 100 can flexibly use network information acquired during a period including a period during which the financial information is aggregated. It is possible to respond.

  On the other hand, the calculation apparatus 100 may generate a model based on the first data when the financial score is calculated and the second data aggregated at an arbitrary time. As a result, the calculation device 100 can generate a model that reflects the reaction, evaluation, and the like on the Internet with respect to a general user's business, regardless of the timing of aggregation.

  Moreover, the calculation apparatus 100 is based on the 1st data at the time of calculating a provider's financial score, the 1st data of the present time estimated based on transition of 1st data, and 2nd data. A model may be generated. In other words, the calculation device 100 predicts the first data at the present time for a predetermined business operator based on the transition of the past first data. Then, the calculation device 100 generates a model using the predicted first data and the second data tabulated at the present time. Thereby, the calculation apparatus 100 can generate | occur | produce the model which calculates the score of a provider with higher freshness.

  In this case, the calculation device 100 is based on the difference between the second data collected at the time corresponding to the first data when the financial score of the operator is calculated and the second data at the present time. The first data at the current time may be predicted. That is, since the calculation device 100 can acquire information on the second data in real time, the difference between the second data when the financial score of the operator is calculated and the second data at the present time Can be requested. The calculation device 100 obtains the correlation between the first data and the second data from the difference. Then, the calculation device 100 predicts the current first data based on the obtained correlation. Thereby, the calculation device 100 can acquire the first data based on the prediction with high reliability. Therefore, the calculation apparatus 100 can generate a model using highly reliable data. Note that the calculation apparatus 100 can obtain the correlation between the first data and the second data based on any known statistical method.

[4-3. Use of related party information)
In the above-described embodiment, the calculation apparatus 100 has shown an example in which a model is generated by using information related to a business manager's manager or officer's name recognition, personal connections, and the like. Here, for example, the calculation apparatus 100 may acquire information from an SNS used by a business owner or an individual officer, and use the information for generating a model.

  For example, the acquisition unit 131 acquires information on purchasing behaviors of business owners and officers from the SNS. In this case, the acquisition unit 131 is that the management or the individual of the manager or the individual of the officer has purchased a relatively expensive product, etc., or has performed frequent investment activities. Get information such as being. And the production | generation part 132 uses the acquired management information of a management or an officer individual as information which digitized the network information of the said provider for the production | generation of a model. For example, the generation unit 132 determines that the management state of the business operator is better as the frequency of purchasing activities or investment behaviors of the manager or individual officer increases, and determines a higher value for the information.

  In addition, the acquisition unit 131 may acquire information on an individual (an individual who is connected in the SNS) that is assumed to have a connection in an SNS that is used by a business manager or an individual officer. For example, the acquisition unit 131 may provide information such as the status of individuals who are connected to the managers and officers of the business operator, the size and management status of the company with which the individual is involved, the name and personality of the individual, and the status of the individual's previous job. To get. The generation unit 132 determines the personality of the business manager and the individual officer based on the number of persons connected to the business manager and the individual officer, and the above-described information on each person. And the production | generation part 132 digitizes network information based on the magnitude of the network power of the manager of the business operator who determined, and the officer individual. In this way, the calculation device 100 can increase the business operator's growth potential by determining the human network power of the business operator, and generate a model based on the growth potential.

  Moreover, the acquisition part 131 may acquire the information regarding an operator's personnel trend among the information transmitted in SNS which the manager of a company and an individual officer use. For example, the acquisition unit 131 acquires information related to job offers and job separations at business operators. And the production | generation part 132 determines with the business continuity of a business operator being opaque when the information regarding the job offer and the job separation in the business is frequently transmitted, and reduces the numerical value of the information. On the other hand, the generation unit 132 improves the numerical value of such information as promising in the growth potential of the operator when the information on the job offer and the expansion of the operator's scale are observed in the long term.

  In addition, for the information transmitted on the Internet, such as SNS, the acquisition unit 131 registers, for example, a word that is assumed to be an evaluation index in advance and automatically collects information from the SNS. Can do. In addition, the acquisition unit 131 can acquire information that can be used as an evaluation index with high accuracy by updating a word registered in advance by machine learning.

[4-4. Profession〕
In the above-described embodiment, the calculation apparatus 100 may generate a model using information related to product trends of peers. For example, the generating unit 132 determines that the scale of the industry as a whole is increasing, the needs of consumers are increasing, and the like from information related to the management status of the same company. Specifically, the generation unit 132 determines that the attention level of the entire industry has improved due to an increase in the number of searches related to peers, an increase in the number of browsing websites, and the like. In this case, the generation unit 132 sets the variables in the regression analysis, digitizes the information based on the improvement in the number of searches related to the same company, the improvement in the number of browsing of the website, and the like. A model that improves the score of a person can be generated.

[4-5. Amount of information)
In the above embodiment, the calculation apparatus 100 has shown an example in which a model is generated based on various information that can be acquired on the network. Here, the calculation apparatus 100 may perform processing using only information acquired from general users who use various websites that exceeds a certain threshold.

  For example, reviews regarding products on a product evaluation site, user evaluations, and the like may show a biased tendency if they are not based on a certain number or more of data. In this case, the calculation device 100 may generate a model that cannot accurately calculate the creditworthiness of the business operator due to the influence of data with a tendency biased toward the regression equation. For this reason, the calculation apparatus 100 can use only data whose reviews and user evaluations transmitted from the user exceed a certain number as data handled in the model calculation process. Thereby, the calculation apparatus 100 can generate a model for calculating a highly reliable score.

[4-6. weight〕
The calculation apparatus 100 may weight specific information for the acquired network information. For example, the calculation device 100 determines that a website or the like where a specialist in a specific field reviews a product has higher reliability than an evaluation site or the like that accepts posts from general users. In other words, the calculation apparatus 100 weights the information related to product reviews and user evaluations on websites where experts review products, more heavily than information that can be acquired from other general sites, and performs model calculation processing. Can be used. Thereby, the calculation apparatus 100 can generate a model for calculating a highly reliable score.

[4-7. correction〕
The calculation apparatus 100 may generate a model that can correct the output score according to the actual economic state. For example, the calculation apparatus 100 classifies a business operator whose business condition is improved when the yen is strong, a business operator who is not affected, and a business person whose business condition is deteriorated. Then, when calculating the operator's score, the calculation apparatus 100 inputs the trend of the value of the yen in a predetermined period to the model, thereby correcting the operator's score corrected by taking into account the trend of the value of the yen. Generate a model that outputs. Such correction can be reflected in a model to be generated by, for example, acquiring business data in the long term and accumulating data related to the trend of the value of the yen.

[4-8. Information on communication network)
In the above-described embodiment, the calculation apparatus 100 has been described in detail with respect to an example in which information based on user behavior is mainly acquired from information on a communication network and related to a business operator. However, the information on the communication network acquired by the calculation device 100 is not limited to the above example.

  For example, the calculation apparatus 100 may acquire information that does not involve the actions of general users who use the communication network as information on the communication network related to the business operator. For example, the calculation apparatus 100 may acquire information related to natural phenomena such as weather information. As an example, the calculation device 100 includes, among weather information and disaster information that can be acquired on the network, weather information and disaster information of a region where the operator is located, or weather information of a region included in the name of the operator, Obtain disaster information as information related to the business. This is because there is a possibility that the business condition of the business operator will fluctuate in the future depending on the weather situation or disaster situation of the area where the business operator is located. By including the element derived from the weather information in this way, the calculation device 100 can generate a model that calculates the creditworthiness evaluation of the operator with higher accuracy.

  In addition, the calculation apparatus 100 is not limited to information transmitted from the user himself / herself, but information (so-called mono-things) that a communicable device located around the user uploads the state of the device on a communication network using a sensor or the like. Information that does not involve the user, such as information on the Internet (Internet of Things), may be acquired. As an example, the calculation device 100 is information transmitted from a product provided by a predetermined provider out of information that can be acquired on a network, and a large number of products are always operating across a wide area. Get information that This is because the penetration rate and occupancy rate of products provided by a business can be an indicator of the business stability of that business. The calculation device 100 is not limited to information transmitted from the user in this way, and includes a component of information transmitted from various objects, thereby generating a model for calculating the creditworthiness evaluation of the operator with higher accuracy. Can do. As described above, the calculation apparatus 100 is not limited to information directly or indirectly related to the user, and generates a model based on various information existing on the communication network. It is possible to generate a highly versatile model applicable to the above.

[4-9. Information on business operators)
The acquisition unit 131 may acquire information other than financial information as information regarding the business acquired through the business data server 50. For example, the acquisition unit 131 appropriately acquires information uniquely collected by the provider data provider. For example, the acquisition unit 131 acquires information related to a business owner. Information on business owners includes, for example, the business manager's name, background, asset security, as well as the management's personality and management ability, as determined from an artificial perspective. It may be information.

[4-10. Information variations)
In the above-described embodiment, the calculation apparatus 100 illustrates an example in which processing is performed using various types of information stored in the network information storage unit 122. Here, the variation of information handled by the calculation apparatus 100 will be described in more detail.

  For example, the calculation device 100 may acquire more detailed information regarding words stored in the word information table 125. For example, the calculation device 100 may perform morphological analysis on the acquired word and acquire information aggregated for each part of speech. Specifically, the calculation apparatus 100 acquires a totaling result obtained by classifying noun categories into people, things, places, and the like. In addition, the calculation device 100 can acquire a classification result such as whether each word has a positive attribute or a negative attribute, and a determination result such as an intention determination of the dialog when the message is in an interactive format.

  Moreover, regarding the acquisition of the message, the calculation device 100 is not limited to the message transmitted when the user uses the search site or the SNS site, and may acquire, for example, the news transmitted by the news site as the message. In addition, the calculation apparatus 100 may acquire a text included in a web page including the business name (for example, a site introducing the business or a site criticizing the business) as a message. Further, the calculation device 100 may acquire a message posted on a finance-related bulletin board or the like. Moreover, the calculation apparatus 100 may acquire voice information transmitted from the user as a message. In other words, the calculation apparatus 100 can acquire text information including words by recognizing a voice transmitted from a user who uses voice search or the like.

  Then, the calculation device 100 generates a model for predicting the operator's score based on data obtained by digitizing the acquired information (in other words, data digitized as word importance). For example, the calculation device 100 divides a message related to a business operator into words by performing morphological analysis, and assigns each word with information quantified by an algorithm such as tf-idf. And the calculation apparatus 100 can utilize the word used regarding a provider for the score calculation of the said provider by using the digitized information as one of the composition of a model.

  Note that the calculation apparatus 100 may extract only a specific part of speech for the acquired word and use it for the score calculation process. For example, the calculation device 100 uses a part of speech that can express the situation of the business operator, such as a noun or an adjective, for processing. Thereby, the calculation apparatus 100 can reduce the processing load without reducing the accuracy of score calculation.

  Moreover, the calculation apparatus 100 may use only words registered in advance for the score calculation process. For example, the calculation apparatus 100 uses words that can directly represent the status of the business operator, such as “bankruptcy”, “growth”, “good”, “bad”, and the like. Such registration is performed by an administrator of the calculation apparatus 100, for example.

  In addition to classifying words as parts of speech, the calculation device 100 may classify the words for each predetermined category. For example, the calculation apparatus 100 registers categories such as “person name”, “location”, and “amount” in advance, and uses information related to words corresponding to the category for processing.

  Further, the calculation apparatus 100 may classify words into positive attributes or negative attributes, and use words determined to have positive attributes or negative attributes for processing. As an example, the calculation apparatus 100 counts the number of words determined to have a positive attribute and the number of words determined to have a negative attribute in one message. Then, the calculation device 100 divides the number of positive attribute words by the number of negative attribute words, and calculates a numerical value indicating the ratio of the two. Thereby, the calculation device 100 can use information such as whether the operator is positively captured or negatively captured by the user in the acquired message as a numerical value.

  Furthermore, the calculation apparatus 100 may acquire a change rate of information regarding the information related to the words described above, and may use it for processing. As an example, the calculation apparatus 100 acquires the change rate of the number of occurrences of “bankruptcy” with respect to the total number of words among all messages related to a predetermined business operator. And the calculation apparatus 100 utilizes the information regarding the acquired change rate for a process of calculating a company's score. Thereby, the calculation apparatus 100 can generate a model that reflects the tendency of general users to use the word “bankruptcy” for a predetermined business operator. As a result, the calculation device 100 can calculate a highly accurate score that reflects the general user's reaction and interest in the creditworthiness of the operator.

  Note that the variation of information as described above is an example, and the calculation apparatus 100 can execute the same processing as the processing described above even if the information is information other than words. For example, the calculation apparatus 100 may acquire a change rate of evaluation of a product provided by the business operator and use the change rate for the processing.

[4-11. Others]
In addition, among the processes described in the above embodiment, all or part of the processes described as being automatically performed can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the document and drawings can be arbitrarily changed unless otherwise specified.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured.

  For example, the information in the storage unit 120 illustrated in FIG. 2 may be stored in an external storage server or the like without the calculation device 100. In this case, the calculation device 100 obtains various stored information by accessing the storage server.

  In addition, for example, the calculation device 100 described above receives a proposal regarding the credit information of the operator or notifies the credit information of the operator, for example, a front-end server side that mainly executes an exchange with an external device, and the Internet It may be distributed to the back-end server side that acquires the above information or executes a calculation process or the like. In this case, for example, the front-end server includes at least a reception unit 133 and a notification unit 135. Further, the back-end server has at least a generation unit 132.

[5. Hardware configuration)
Further, the calculation apparatus 100 according to the embodiment described above is realized by a computer 1000 having a configuration as shown in FIG. 14, for example. FIG. 14 is a hardware configuration diagram illustrating an example of a computer 1000 that implements the functions of the calculation apparatus 100. The computer 1000 includes a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

  The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400 and controls each unit. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started up, a program depending on the hardware of the computer 1000, and the like.

  The HDD 1400 stores a program executed by the CPU 1100, data used by the program, and the like. The communication interface 1500 receives data from other devices via the communication network 500 (corresponding to the communication network according to the embodiment) and sends the data to the CPU 1100, and other data generated by the CPU 1100 via the communication network 500. To the device.

  The CPU 1100 controls an output device such as a display and a printer and an input device such as a keyboard and a mouse via the input / output interface 1600. The CPU 1100 acquires data from the input device via the input / output interface 1600. Further, the CPU 1100 outputs the data generated via the input / output interface 1600 to the output device.

  The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads such a program from the recording medium 1800 onto the RAM 1200 via the media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase change rewritable disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

  For example, when the computer 1000 functions as the calculation device 100, the CPU 1100 of the computer 1000 implements the function of the control unit 130 by executing a program loaded on the RAM 1200. In addition, each data in the storage unit 120 is stored in the HDD 1400. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired from other devices via the communication network 500.

[6. effect〕
As described above, the calculation device 100 according to the embodiment includes the first data that is information related to the financial status of the business operator and the information related to the business operator, and the user device on the communication network. An acquisition unit 131 that acquires second data that is information based on the first data and the second data acquired by the acquisition unit 131, and a score (index value) indicating the creditworthiness of the operator. And a calculating unit 134 for calculating.

  As described above, the calculation device 100 according to the embodiment calculates a score indicating the creditworthiness of the operator based on not only the financial information generally used for evaluating the creditworthiness of the operator but also the user behavior in the network. That is, since the calculation apparatus 100 can calculate the reliability based on a large amount of information such as general user behavior information, the calculation apparatus 100 can calculate a highly accurate reliability. In addition, the calculation apparatus 100 can generally calculate the creditworthiness suitable for current affairs by using, in the calculation process, network information with high freshness that is constantly updated together with financial information updated every quarter or the like.

  In addition, the calculation apparatus 100 according to the embodiment further includes a generation unit 132 that generates a model for predicting a score indicating the creditworthiness of the operator based on the first data and the second data. Further, the calculation unit 134 calculates an index value indicating the creditworthiness of the operator by using the model generated by the generation unit 132.

  As described above, the calculation apparatus 100 calculates the operator's score based on the model that receives the network information as well as the financial information. Thereby, since the calculation apparatus 100 can output the operator's score immediately by inputting the information of the operator to be processed, the calculation device 100 can calculate a high-accuracy credibility, The burden can be reduced.

  In addition, the acquisition unit 131 acquires the financial score (an example of “evaluation value”) of the business operator calculated in advance based on the first data. The generation unit 132 generates a model based on the correlation between the financial score acquired by the acquisition unit 131 and the first data and the second data.

  As described above, the calculation device 100 generates a model in which the correlation between the financial score and the network information is reflected based on the correlation between the financial score calculated based on the financial information and the acquired data. . Thereby, the calculation apparatus 100 can calculate the score which shows a provider's reliability with high precision using the model which inputs financial information and network information.

  Moreover, the production | generation part 132 produces | generates a model based on the 1st data at the time of a financial score being calculated, and the 2nd data totaled at the time corresponding to the said 1st data.

  As described above, the calculation apparatus 100 appropriately reflects the correlation between the network information acquired from the general user and the financial information by using the network information at the time corresponding to the financial information. Generate the model. Thereby, the calculation apparatus 100 can calculate the operator's score with high accuracy.

  Moreover, the production | generation part 132 is based on the 1st data at the time of calculating a financial score of a provider, the 1st data of the present time estimated based on transition of the 1st data, and the 2nd data A model may be generated.

  In this way, the calculation device 100 can generate a model for calculating the score of a business operator with high freshness by using the current data.

  In addition, the generation unit 132 determines the current time based on the difference between the second data collected at the time corresponding to the first data when the financial score of the business operator is calculated and the current second data. The first data may be predicted.

  Thus, the calculation apparatus 100 predicts the first data at the present time by using the difference between the second data that is easily acquired in real time as compared with the first data. Thereby, the calculation apparatus 100 can generate a model using highly reliable data.

  In addition, the acquisition unit 131 acquires at least one of the number of searches based on a search query related to the business operator, the search ranking, and the variation rate of the number of searches for each predetermined period as the second data. The generation unit 132 generates a model based on information obtained by digitizing at least one of the number of searches acquired by the acquisition unit 131, the search ranking, and the variation rate of the number of searches for each predetermined period.

  As described above, the calculation device 100 can determine the degree of attention from the general user with respect to the evaluation target business operator by analyzing the search information. Thereby, the calculation device 100 can generate a highly accurate model in which the degree of attention from a general user is one of the determination factors for the creditworthiness of the business operator. Note that, as described above, the calculation device 100 may appropriately use a normalized numerical value in accordance with information to be handled.

  Moreover, the acquisition part 131 acquires at least any one of the browsing number in the website which a provider provides, the number of viewers, and a conversion rate as 2nd data. The generation unit 132 generates a model based on information obtained by digitizing at least one of the number of browsing, the number of viewers, and the conversion rate acquired by the acquisition unit 131.

  Thus, the calculation apparatus 100 can determine the interest from the general user with respect to the evaluation target business operator by analyzing the information on the website provided by the business operator. Thereby, the calculation apparatus 100 can generate a highly accurate model in which the interest from the general user is one of the determination factors for the creditworthiness of the business operator.

  Moreover, the acquisition part 131 acquires the word relevant to a provider as 2nd data. The generation unit 132 generates a model based on the importance of the word related to the business operator acquired by the acquisition unit 131.

  As described above, the calculation device 100 can determine the degree of attention from the general user to the business operator by analyzing the text information posted by the user. Specifically, the calculation device 100 determines the importance of a word related to a business operator (which can be read as a predetermined index value), thereby calculating the creditworthiness of the business operator based on word information on the network. Can be incorporated. Thereby, the calculation apparatus 100 can calculate the creditworthiness of the business operator with high accuracy.

  The acquisition unit 131 also displays the number of words related to the business operator, the number of words that co-occur with the business operator, the characteristics of the words co-occurring with the business operator, Obtain at least one of the weights. The generation unit 132 generates the number of words related to the business acquired by the acquisition unit 131, the number of words that co-occur with the business, the characteristics of the words that co-occur with the business, the business in a predetermined document. Using at least one of the co-occurring word weights, the importance of the word related to the business operator is determined, and a model is generated.

  In this way, the calculation device 100 acquires the number of occurrences of words related to the business operators and the words that co-occur with the keywords related to the business operators so that the impression that the user has in the business operators and the degree of attention to the business operators can be obtained. It can be grasped more accurately. Note that the calculation apparatus 100 can accurately grasp a word that co-occurs with a word related to the operator by using an index value such as a weight of the co-occurring word in a predetermined document. Thereby, the calculation apparatus 100 can generate a highly accurate model in which the degree of attention and impression from the general user is one of the determination factors for the creditworthiness of the business operator.

  The acquisition unit 131 also includes text data transmitted from a user, text data distributed from a news site, text data recognized from voice transmitted from a user, and text data distributed from a website related to a business operator. The word related to the operator is acquired from at least one of the above.

  Thus, the calculation apparatus 100 can acquire words related to the business operator from various routes. Thereby, the calculation apparatus 100 can improve the reliability of data by aggregating many data. Moreover, according to the calculation apparatus 100, it is possible to prevent the data to be processed from being biased toward a specific site or user.

  In addition, the generation unit 132 classifies the words for each part of speech, classifies the words for each predetermined category, classifies the words for each intention (for example, a positive element or a negative element), or the results of these results. Using at least one of the change rates, the importance of the word related to the business operator is determined.

  As described above, the generation unit 132 can determine the importance of the word related to the business operator using various methods for the acquired word. Thereby, the production | generation part 132 can attain diversification of the model for calculating an operator's score.

  In addition, the acquisition unit 131 acquires at least one of the evaluation from the user, the number of users of the product, and the number of reviews posted on the product as the second data. The generating unit 132 converts the information obtained by the acquiring unit 131 into the information obtained by quantifying at least one of the evaluation from the user in the product provided by the operator, the number of users of the product, and the number of reviews posted about the product. Based on this, a model is generated.

  As described above, the calculation apparatus 100 acquires information related to the evaluation of the business operator (or the product provided) from the general user by analyzing the information related to the product provided by the business operator. The calculation apparatus 100 can appropriately indicate the business continuity and growth potential of the business operator as a score by reflecting the evaluation of the product possessed by the business operator. Moreover, since the evaluation from a general user is reflected immediately in the site etc. regarding the evaluation of a product, the calculation apparatus 100 can acquire the reaction with respect to a provider from a general user more directly. For this reason, the calculation device 100 can generate a model that reflects the user's evaluation and can calculate a score that is superior in immediacy.

  In addition, the acquisition unit 131 acquires, as the second data, at least one of the number of downloads of the program product provided by the operator, the number of users, the average usage time per user, or the operation rate during a predetermined period. get. The generation unit 132 quantifies at least one of the number of downloads of the program product acquired by the acquisition unit 131, the number of users, the average usage time per user, or the operation rate during a predetermined period. Based on the above, a model is generated.

  Thereby, the calculation apparatus 100 can determine the business continuity, growth potential, etc. of the business operator as in the product evaluation. Moreover, the calculation apparatus 100 can generate a model in which a response from a real-time user sent to an app store or the like is added to a determination element.

  Further, the acquisition unit 131 acquires at least one of the number of business customers, the continuous usage rate of customers, and the unit price of customers as the second data. The generation unit 132 generates a model based on information obtained by quantifying at least one of the number of business customers, the continuous usage rate of the customer, and the unit price of the customer acquired by the acquisition unit 131.

  Thus, the calculation apparatus 100 can determine the business status of the business operator by analyzing the information related to the business customer. Thereby, since the calculation apparatus 100 can add an operator's bankruptcy probability, business continuity, etc. to a determination element, it can produce | generate the model which calculates a more highly accurate score.

  The acquisition unit 131 acquires first data and second data for businesses belonging to a common industry. The calculation unit 134 calculates an index value indicating the creditworthiness of the operator based on the first data and the second data of the operators belonging to the common industry.

  As described above, the calculation apparatus 100 generates a model using information on business operators belonging to a common industry. Thereby, since the calculation apparatus 100 can generate a model that takes into account similarities such as numbers used in the industry, it can calculate the creditworthiness of the operator with higher accuracy.

  In addition, the acquisition unit 131 acquires the first data via a predetermined platform on which a business operator or a third party can provide information. The calculation unit 134 calculates a score indicating the creditworthiness of the operator based on information provided by the operator or a third party.

  In this way, the calculation device 100 publishes a platform and widely accepts information from the operator itself or a third party, so that the financial information to be disclosed cannot be evaluated normally and the financial information to be disclosed is not sufficient. The credit rating can be calculated with high accuracy even for small and medium enterprises such as companies.

  The embodiments of the present application and some of the modifications of the embodiments have been described in detail with reference to the drawings. The present invention can be implemented in other forms based on various modifications and improvements.

  The above-described calculation device 100 may be realized by a plurality of server computers, and depending on the function, the configuration may be realized by calling an external platform or the like with an API (Application Programming Interface) or network computing. Can be changed flexibly.

  Further, “section (module, unit)” described in the claims can be read as “means”, “circuit”, and the like. For example, the generation unit can be read as generation means or a generation circuit.

DESCRIPTION OF SYMBOLS 1 Calculation system 10 User terminal 30 Financial institution server 40 Web server 50 Operator data server 55 Operator 100 Calculation apparatus 110 Communication part 120 Storage part 121 Operator data storage part 122 Network information storage part 123 Search information table 124 Site information table 125 Word information table 126 Product information table 127 Social information table 128 Customer information table 129 Model storage unit 130 Control unit 131 Acquisition unit 132 Generation unit 133 Reception unit 134 Calculation unit 135 Notification unit

Claims (19)

An acquisition unit that acquires first data that is information related to the financial status of the operator and second data that is information related to the operator and is based on user behavior on the communication network; ,
Based on the first data and the second data acquired by the acquisition unit, a calculation unit that calculates an index value indicating the creditworthiness of the operator,
A calculation device comprising: A generating unit that generates a model for predicting an index value indicating the creditworthiness of the operator based on the first data and the second data;
Further comprising
The calculation unit includes:
By using the model generated by the generation unit, an index value indicating the creditworthiness of the operator is calculated.
The calculation apparatus according to claim 1. The acquisition unit
Obtaining an evaluation value of the operator calculated in advance based on the first data;
The generator is
Based on the correlation between the evaluation value of the operator acquired by the acquisition unit and the first data and the second data, the model is generated.
The calculation apparatus according to claim 2, wherein: The generator is
Based on the first data when the evaluation value of the operator is calculated and the second data aggregated at a time corresponding to the first data, the model is generated.
The calculation apparatus according to claim 3. The generator is
Based on the first data and the second data predicted based on the transition of the first data and the first data when the evaluation value of the operator is calculated, the model is Generate,
The calculation apparatus according to claim 3. The generator is
Based on the difference between the second data collected at the time corresponding to the first data when the evaluation value of the operator is calculated and the current second data, the current first data Predict data,
The calculation device according to claim 5, wherein: The acquisition unit
As the second data, obtain at least one of the number of searches by a search query related to the business operator, a search ranking, and the rate of change in the number of searches per predetermined period,
The generator is
The model is generated based on information obtained by quantifying at least one of the number of searches acquired by the acquisition unit, the search ranking, and the variation rate of the number of searches for each predetermined period.
The calculation device according to any one of claims 2 to 6, wherein The acquisition unit
As the second data, obtain at least one of the number of browsing, the number of viewers, and the conversion rate on the website provided by the business operator,
The generator is
Based on the information obtained by quantifying at least one of the number of browsing, the number of viewers, and the conversion rate acquired by the acquisition unit, the model is generated.
The calculation device according to any one of claims 2 to 7. The acquisition unit
As the second data, obtain a word related to the operator,
The generator is
Generating the model based on the importance of the words related to the operator acquired by the acquisition unit;
The calculation device according to any one of claims 2 to 8. The acquisition unit
At least the number of occurrences of words related to the business operator, the number of occurrences of words co-occurring with the business operator, the characteristics of the words co-occurring with the business operator, and the weight of words co-occurring with the business operator in a predetermined document Get one of them
The generator is
The number of occurrences of the word acquired by the acquisition unit, the number of occurrences of the word co-occurring with the operator, the characteristics of the word co-occurring with the operator, the weight of the word co-occurring with the operator in a predetermined document Determining the importance of words related to the operator using at least one of them, and generating the model;
The calculation device according to claim 9. The acquisition unit
From at least one of text data transmitted from a user, text data distributed from a news site, text data recognized from voice transmitted from a user, and text data distributed from a website related to a business , Get words related to the operator,
The calculation apparatus according to claim 9 or 10, wherein: The generator is
As a result of classifying the words for each part of speech, as a result of classification for each predetermined category, as a result of classification for each intention of the word, or using a change rate of these results, the operator Determine the importance of words related to
The calculation device according to any one of claims 9 to 11. The acquisition unit
As the second data, obtain at least one of an evaluation from a user in a product provided by the business operator, the number of users of the product, and the number of reviews posted about the product,
The generator is
Based on the information obtained by quantifying at least one of the evaluation from the user in the product provided by the provider, the number of users of the product, and the number of reviews posted about the product, acquired by the acquisition unit, Generate a model,
The calculation device according to any one of claims 2 to 12, wherein The acquisition unit
As the second data, obtain at least one of the number of downloads of the program product provided by the business operator, the number of users, the average usage time per user, or the operation rate in a predetermined period,
The generator is
Based on the information obtained by quantifying at least one of the number of downloads of the program product acquired by the acquisition unit, the number of users, the average usage time per user, or the operation rate in a predetermined period, Generate a model,
The calculation device according to claim 2, wherein The acquisition unit
As the second data, obtain at least one of the number of customers of the operator, the continuous usage rate of the customer, and the unit price of the customer,
The generator is
Generating the model based on information obtained by quantifying at least one of the number of customers of the operator acquired by the acquisition unit, the continuous usage rate of the customer, and the unit price of the customer;
The calculation device according to any one of claims 2 to 14, wherein The acquisition unit
For businesses belonging to a common industry, the first data and the second data are acquired,
The calculation unit includes:
Based on the first data and the second data of businesses belonging to a common industry, an index value indicating the creditworthiness of the businesses is calculated.
The calculation device according to claim 1, wherein The acquisition unit
The first data is acquired through a predetermined platform on which the operator or a third party can provide information,
The calculation unit includes:
The calculation device according to any one of claims 1 to 16, wherein an index value indicating creditworthiness of the operator is calculated based on information provided by the operator or a third party. A calculation method executed by a computer,
An acquisition step of acquiring first data that is information related to the financial status of the operator and second data that is information related to the operator and is based on user behavior on the communication network; ,
A calculation step of calculating an index value indicating the creditworthiness of the operator based on the first data and the second data acquired by the acquisition step;
The calculation method characterized by including. An acquisition procedure for acquiring first data that is information related to the financial status of the operator and second data that is information related to the operator and is based on user behavior on the communication network; ,
A calculation procedure for calculating an index value indicating the creditworthiness of the operator based on the first data and the second data acquired by the acquisition procedure;
A calculation program for causing a computer to execute.

Images