WO2020240714A1

WO2020240714A1 - Search system, search method, and search application software

Info

Publication number: WO2020240714A1
Application number: PCT/JP2019/021156
Authority: WO
Inventors: 宇雄國井; 隆徳青井; 悦至喜多村
Original assignee: リンカーズ株式会社
Priority date: 2019-05-28
Filing date: 2019-05-28
Publication date: 2020-12-03
Also published as: JP6890671B2; JPWO2020240714A1

Abstract

[Problem] To provide a search system, a search method, and a search application software that make it possible to easily search for technical information that is of high interest to a user in an interactive manner, even for searches for technical information for which selection of appropriate keywords is difficult. [Solution] A search system that can be executed by a search server 20 that can communicate with a user terminal 10 into which a user inputs search conditions. A plurality of pieces of information about a first technology are extracted from a technical information database 24b at the search server 20 that stores information about a plurality of technologies, and the information about the first technology is evaluated at the user terminal 10. On the basis of evaluation results for the information about the first technology as evaluated at the user terminal 10, a relevance score calculation unit 27 calculates a relevance score that is the result of performing prescribed processing on the similarity between the plurality of pieces of information about the first technology and information about a second technology that is stored at the technical information database 24b. The information about the plurality of technologies that is stored at the technical information database 24b is rearranged on a user by user basis in order of highest relevance score and stored with relevance scores at a technical information list database 24d. Information about a prescribed technology is extracted from the technical information list database 24d on the basis of search conditions.

Description

Search system, search method and search application software

The present invention relates to a search system, a search method, and search application software for searching information related to technologies such as documents, human resources, and companies.

Conventionally, there are a plurality of search engines for searching for desired information from a plurality of pieces of information. Many conventional search engines require the user to enter a keyword related to the information to be searched, and search for information including the keyword entered by the user. Further, regarding the search for specific technical information such as patented technology, technical classification information such as international patent classification may be used in addition to the keyword, but in such a case, the search by keyword is indispensable.

In this way, keyword search is an important means for the user who searches for information on technology to reach the information of most certainty, regardless of the type of information on technology. However, in general, it is difficult for a user to obtain information of interest from a huge amount of search results by assigning appropriate keywords, and in order to obtain desired information from various information sources, the keywords are input by trial and error. In addition, there is a problem that it takes a considerable amount of time to input and it depends largely on the experience and intuition of the user who performs the search.

Therefore, as in the invention described in Patent Document 1, in order to obtain an associative relationship with the input word, a search for another word related to the input word is performed in a plurality of word inputs by the user. There is a search system that outputs words or word information existing between input words based on the word space generated by repeating.

Japanese Unexamined Patent Publication No. 2012-123369

However, the invention described in Patent Document 1 is appropriate when the user cannot select an appropriate keyword because the search result presented to the user is limited to the words input by the user himself / herself. It is difficult to obtain new search results, and it is also difficult to obtain new information that the user never expected. Further, the search act is merely a repetition of the result output of the keyword input, and the user is not devised so that the search act itself can be performed with interest.

The present invention has been made in view of such circumstances, and even when searching for technical information for which it is difficult to select an appropriate keyword, the user can easily interactively search for desired technical information while enjoying the search action itself. An object of the present invention is to provide a search system, a search method, and search application software that can be acquired.

In order to achieve the above object, the search system of the present invention connects a user terminal for a user to input a search condition to the user terminal via a database, and searches for information based on the search condition. A search system including a server, wherein the search server extracts a technical information database for storing information on a plurality of technologies and information on a plurality of first technologies to be evaluated by the user terminal from the technical information database. Based on the evaluation results of the first technical information extraction unit and the information related to the plurality of first technologies in the user terminal, the information related to the plurality of first technologies and the second technical information database are stored. The relevance score calculation unit that calculates the relevance score obtained by performing predetermined processing on the similarity with the information related to the technology, and the information about the plurality of technologies stored in the technical information database are stored in the technical information database for each user. From the technical information list generation unit that generates a list sorted in descending order of score, the technical information list database that stores the list generated by the technical information list generation unit, and the technical information list database based on the search conditions. It is characterized by including a second technical information extraction unit that extracts information about a predetermined technology.

The first technical information extraction unit and the second technical information extraction unit are realized by a control unit, a storage unit, and a communication unit provided in the search server according to the present embodiment described below. It means that the control unit extracts information about a predetermined technique stored in the storage unit based on a predetermined condition, and the communication unit transmits the information to the user terminal.
Further, since the same applies to the search method and the search application software, the description thereof will be omitted.

Further, in the search system of the present invention, the relevance score calculation unit calculates the similarity between the information related to the plurality of first techniques and the information related to the second technique, and determines the similarity with the plurality of first techniques. It is characterized in that an adjusted score adjusted based on the evaluation result for each of the information related to the technique 1 is calculated, and the calculated plurality of adjusted scores are added up to calculate the relevance score.

Further, in the search system of the present invention, the relevance score calculation unit is at least several times or more the adjusted score of the information on the first technology and the information on the second technology highly evaluated by the user. A positive number, or a negative number that is at least several times or more the adjusted score of the information about the first technology and the information about the second technology that the user has evaluated low, is the information about the first technology. It is characterized by being calculated as a relevance score.

Further, in the search system of the present invention, when the relevance score calculation unit is expressed by a numerical value in the range of -1 to +1 for the similarity, the information regarding the first technique in which the evaluation result is highly evaluated. The adjusted score is calculated by multiplying or adding a coefficient of 1.0 or more according to the value of the similarity between the above and the information regarding the second technique, and the evaluation result is low. It is characterized in that the adjusted score is calculated by performing adjustment by multiplying or subtracting a coefficient of 0 or less according to the value of the similarity between the information on the first technique and the information on the second technique. ..

Further, in the search system of the present invention, the relevance score calculation unit determines the magnitude of the similarity based on the magnitude of the similarity between the information regarding the first technique and the information regarding the second technique. It is characterized in that the adjustment degree score is calculated by performing the adjustment in multiple steps.

Further, in the search system of the present invention, the search server includes a user database for storing information about a technique of interest to the user as interest information for each user, and the evaluation unit is stored in the user database. It is characterized in that information on the plurality of techniques for being evaluated by a user is extracted from the technical information database based on the above-mentioned interest information.

Further, in the search system of the present invention, the search server has a first extraction unit that extracts information about a predetermined technology from the technical information list database that stores information about the technology evaluated within a predetermined period, and a past It is provided with a second extraction unit that extracts information on a predetermined technology from the technical information list database that stores information on the technology that was last evaluated without considering the evaluation, and the technology is related by either extraction unit. It is characterized in that it is possible to select to extract information.
The first extraction unit and the second extraction unit are realized by a control unit, a storage unit, and a communication unit provided in the search server according to the following embodiment, and specifically, under predetermined conditions. It means that the control unit extracts information about a predetermined technique stored in the storage unit based on the information, and the communication unit transmits the information to the user terminal.
Further, since the same applies to the search application software, the description thereof will be omitted.

Further, in the search system of the present invention, the user terminal displays information on the plurality of first technologies extracted by the first technical information extraction unit one by one, and causes the user to determine whether or not he / she is interested. The information regarding the first technology determined to be of interest to the user is regarded as highly evaluated information, and the information regarding the first technology determined to be uninterested by the user is regarded as low evaluated information. It is a feature.

Further, the search system of the present invention is a calculation method or a calculation method used for processing of the relevance score calculation unit and the technical information list generation unit based on the evaluation result regarding the information related to the first technology evaluated by the user terminal. It is characterized in that parameters can be selected.

Further, the search method of the present invention is a search method by a search server that connects a user terminal into which a user inputs a search condition to the user terminal via a network and searches for information based on the search condition. The first step of extracting information on a plurality of first technologies to be evaluated by the user terminal from a technical information database that stores information on a plurality of technologies included in the search server, and the evaluation by the user terminal. Based on the evaluation results of the information on the plurality of first technologies, a predetermined process was performed on the degree of similarity between the information on the plurality of first technologies and the information on the second technology stored in the technical information database. The second step of calculating the relevance score and the information about the plurality of technologies stored in the technical information database are sorted for each user in descending order of the relevance score, and the technical information is sorted together with the relevance score. It is characterized by having a third step of storing in a list database and a fourth step of extracting information about a predetermined technique from the technical information list database based on the search condition.

Further, the search application software of the present invention is search application software that can be executed in a search server that searches for information based on the search conditions and can communicate with a user terminal that allows the user to input search conditions. A first function for extracting information on a plurality of first technologies for evaluation by the user terminal from a technical information database for storing information on a plurality of technologies provided in the user terminal, and the plurality of first functions evaluated on the user terminal. Based on the evaluation result of the information related to the first technology, the relevance score obtained by performing a predetermined process on the similarity between the information related to the plurality of first technologies and the information related to the second technology stored in the technical information database. And the information about the plurality of technologies stored in the technical information database are sorted for each user in descending order of the relevance score, and are stored in the technical information list database together with the relevance score. The third function to memorize and
A fourth function of extracting information about a predetermined technology from the technical information list database based on the search condition can be executed by the search server.

Further, in the search application software of the present invention, the second function calculates the similarity between each of the information related to the plurality of first technologies and the information related to the second technology, and the similarity is used as the first. It is characterized in that an adjusted score adjusted based on the evaluation result for each of the information related to the technology of 1 is calculated, and the calculated adjusted scores are added up to calculate the relevance score.

Further, in the search application software of the present invention, the second function is at least several times or more the adjusted score of the information on the first technology and the information on the second technology highly evaluated by the user. A positive number or a negative number that is at least several times or more the adjusted score of the information about the first technology and the information about the second technology that the user has given a low rating is the relevance score of the first information. It is characterized by calculating as.

Further, in the search application software of the present invention, when the second function is expressed by a numerical value in the range of -1 to +1 for the similarity, the information regarding the first technique in which the evaluation result is highly evaluated. The adjusted score is calculated by multiplying or adding a coefficient of 1.0 or more according to the value of the similarity between the above and the information regarding the second technique, and the evaluation result is low. It is characterized in that the adjusted score is calculated by performing adjustment by multiplying or subtracting a coefficient of 0 or less according to the value of the similarity between the information regarding the first technique and the information regarding the second technique.

Further, in the search application software of the present invention, the second function has the magnitude of the similarity based on the magnitude of the similarity between the information regarding the first technique and the information regarding the second technique. It is characterized in that the adjustment degree score is calculated by performing the adjustment in multiple steps.

Further, the search application software of the present invention can further execute a fifth function of storing information of interest regarding the user's technology in the user database provided in the search server, and the first function is provided in the user database. It is characterized in that information on the plurality of first technologies is extracted from the technical information database that stores information on the plurality of technologies included in the search server based on the stored matters of interest.

Further, the search application software of the present invention does not consider the first mode of extracting information on a predetermined technology from the technical information list database that stores information on the technology evaluated within a predetermined period, and past evaluations. It is characterized in that a sixth function of selecting a second mode for extracting information on a predetermined technology from the technical information list database for storing information on the last evaluated technology is further feasible. ..

Further, in the search application software of the present invention, the first function causes the user terminal to display information on the plurality of first technologies one by one to determine whether or not the user is interested, and the user is interested. The information relating to the first technology determined to be present is regarded as highly evaluated information, and the information relating to the first technology determined to be uninterested by the user is regarded as low evaluation information.

Further, the search application software of the present invention is a calculation method or a calculation method used for processing the second function and the third function based on the evaluation result regarding the information regarding the first technology evaluated on the user terminal. The feature is that parameters can be selected.

According to the present invention, even when searching for technical information for which it is difficult to select an appropriate keyword, it is possible to easily search for technical information that is of great interest to the user in an interactive format.

Also, even if the user's interest in technology changes, or if he / she is interested in multiple technologies at the same time, it is possible to set and adjust multiple search methods.

It is a figure which shows the structure of the search system which concerns on one Embodiment of this invention. It is a block diagram which shows the structure of the user terminal which is one component of the search system which concerns on this embodiment. It is a block diagram which shows the structure of the search server which is one component of the search system which concerns on this embodiment. It is a block diagram which shows the detailed structure of the storage part of the search server which concerns on this embodiment. It is a figure which shows the structure of the user database stored in the storage part which concerns on this embodiment. It is a figure which shows the structure of the document information stored in the technical information database provided in the storage part which concerns on this embodiment. It is a figure for demonstrating the index which defined the degree of progress of the technical development about the individual literature information used in the search system which concerns on this Embodiment in a multi-step. It is a figure which shows the structure of the matter information stored in the technical matter information database provided in the storage part which concerns on this embodiment. It is a figure which shows the structure of the matter information stored in the technical information list database provided in the storage part which concerns on this embodiment. It is a figure which shows the structure of the user evaluation information stored in the user evaluation log database provided in the storage part which concerns on this embodiment. It is a figure for demonstrating an example of the calculation logic performed by the relevance score calculation part concerning this embodiment. It is a flowchart for demonstrating the method of setting the technical interest information of a user in the user terminal of the search system which concerns on one Embodiment of this invention. It is a flowchart for demonstrating the method of setting the technical interest information of a user in the search server of the search system which concerns on one Embodiment of this invention. It is a sequence diagram which shows an example of the setting procedure of the technical interest information in the search system which concerns on one Embodiment of this invention. It is a flowchart for demonstrating the search process of the technical information in the user terminal of the search system which concerns on one Embodiment of this invention. It is a flowchart for demonstrating the search process of the technical information in the search server of the search system which concerns on one Embodiment of this invention. It is a sequence diagram which shows an example of the document search processing procedure in the search system which concerns on one Embodiment of this invention. It is a figure which shows an example of the mode selection screen displayed on the user terminal at the start of the document search process in the search system which concerns on one Embodiment of this invention. It is a figure which shows an example of the search information setting screen in the express mode which is displayed on the user terminal which concerns on this embodiment. It is a figure which shows an example of the technical evaluation screen displayed on the display part of a user terminal when a user performs evaluation work. It is a figure which shows an example of the filter setting screen displayed on the display part of a user terminal. It is a figure which shows an example of the search condition setting change screen displayed on the display part of a user terminal. It is a figure which shows an example of the screen which prompts the user's instruction input displayed on the display part of a user terminal. It is a flowchart for demonstrating the detail of the output process (step S409) of the technical information list of the search server in this embodiment.

Hereinafter, the search system, the search method, and the search application software according to the embodiment of the present invention will be described with reference to the drawings.
In the present embodiment, documents such as journals of technical societies are assumed as an example for searching information on technology, and information on these documents and information on technology described in the documents (hereinafter referred to as "technical information") are searched. The processing will be described.

<Search system 1>
First, a search system according to an embodiment of the present invention will be described.
FIG. 1 is a diagram showing a configuration of a search system according to an embodiment of the present invention.
As shown in the figure, in the search system 1, a plurality of user terminals 10 and a search server 20 are connected via a network 30. The user terminal 10 is assumed to be a multifunctional mobile phone such as a smartphone owned or assigned to a developer or researcher of a private company, a researcher or a student belonging to a university or a public research institution, etc., but mobile information It may be a terminal (PDA), a personal computer, or the like.

The search server 20 is a server for accumulating information on technology such as research papers and product catalogs in each technical field and providing a search service, and may be a shared server, a dedicated server, a personal computer, a public cloud, or the like. You may.

In the embodiment, the search server 20 stores information about the technology, and although the processing in the illustrated search system 1 is closed, the technology is related to the database (not shown) outside the search system 1. The process of downloading information and storing it in the search server 20 may be appropriately performed before and after the search process or during the search process.

In addition, when using data from an external database, even if the data is appropriately processed and stored in a format that can be executed by the search process according to the present embodiment, the data is stored as it is and the search process is performed. Sometimes you may want to retrieve only the data you need.

The network 30 connects the user terminal 10 and the search server 20 to each other, and is a LAN or the like inside an organization such as the Internet or inside a company or university. Further, the search application software (program) according to the present invention is executed on the user terminal 10 and the search server 20. When executing these programs, communication is performed using a communication protocol such as Hypertext Transfer Protocol (HTTP). In the present embodiment, the user terminal 10 and the search server 20 are described as a wirelessly connected network system, but a part or all of the network system may be connected by wire.

<User terminal 10>
FIG. 2 is a block diagram showing a configuration of a user terminal 10 which is a component of the search system according to the present embodiment.
As shown in the figure, the user terminal 10 according to the present embodiment includes a control unit 11, an operation unit 12, a display unit 13, a storage unit 14, and a communication unit 15. The control unit 11 performs various arithmetic processes for controlling the entire function of the user terminal 10 by the program instruction group and the software instruction group stored in the storage unit 14.

The operation unit 12 receives an input operation by the user, and is composed of, for example, a touch panel and operation buttons. The display unit 13 displays an image, characters, or the like on a liquid crystal display or an organic EL display on the outer surface of the user terminal 10 based on a command from the control unit 11.

The storage unit 14 stores various programs and various data downloaded by pre-installation or via the network 30. The communication unit 15 includes an antenna, a transmission / reception circuit, and the like, and transmits / receives various data to / from the search server 20 via the network 30.

<Search server 20>
FIG. 3 is a block diagram showing a configuration of a search server 20 which is a component of the search system according to the present embodiment.
As shown in the figure, the search server 20 includes a control unit 21, an operation unit 22, a display unit 23, a storage unit 24, a communication unit 25, a technical information list generation unit 26, and a relevance score calculation unit 27. To be equipped.

The control unit 21 performs various arithmetic processes and the like for controlling the functions of the entire search server 20. The operation unit 22 is realized by, for example, a touch panel, a keyboard, a mouse, or the like, and receives an input operation of an administrator who operates and manages the search server 20. The display unit 23 displays information such as images and characters based on the command of the control unit 21. The storage unit 24 stores various data such as an initial program for starting the search server 20, a program for performing a search service, and information on a technology to be searched by the user terminal 10.

The communication unit 25 controls for sending and receiving data to and from the user terminal 10 and an external database. The technical information list generation unit 26 searches for information on the technology stored in the storage unit 24 based on the information input by the user terminal 10 and generates technical information such as a list information of technical documents.

The "first technical information extraction unit" and the "second technical information extraction unit" according to claim 1 of the claims include the control unit 21 included in the search server 20 according to the present embodiment. It can be realized by the storage unit 24 and the communication unit 25. Specifically, the control unit 21 extracts information about a predetermined technique from the storage unit 24 based on a predetermined condition, and the communication unit 25 extracts the information. It means transmitting information about a predetermined technique to the user terminal 10.

The relevance score calculation unit 27 can correlate with the similarity between information related to a plurality of technologies such as a plurality of technical documents stored in the storage unit 24 and the interest of the user input by the user on the user terminal 10. The relevance score is calculated based on user input information such as parameters.

Next, the detailed configuration of the storage unit 24 will be described.
FIG. 4 is a block diagram showing a detailed configuration of the storage unit 24 of the search server 20 according to the present embodiment.
As shown in the figure, the storage unit 24 has a user database 24a that stores information (user information) about the user of the search system 1 according to the present embodiment, and a technical information database 24b that stores information about various technologies such as technical document information. A search matter information database 24c that stores information for each matter related to the search processing of information related to the technology performed by the user, and a technical information list database 24d that stores the technical information list created by the result of the user's search processing. It is provided with a user evaluation log database 24e that stores evaluation information of information related to technology by the user.
Further, the technical information list database 24d stores information such as the clustering result 24d4 and the relevance score 24d6 calculated by the relevance score calculation unit 27 as described later.

Next, the details of the various databases 24a to 24e included in the storage unit 24 will be described.

(A) User database 24a
FIG. 5 is a diagram showing a configuration of a user database 24a stored in the storage unit 24 according to the present embodiment.
As shown in the figure, the user database 24a includes a user ID 24a1 assigned to each user, user terminal information 24a2 regarding the user terminal 10 used by the user, information 24a3 necessary for controlling distribution information to the user terminal 10, and the user. It stores technical interest information 24a4 such as keywords related to technical interest set during character calculation or search.

As shown in the figure, in the present embodiment, one user ID is assigned to each of a maximum of 9999 users from 0001 to 9999 so that a favorite search can be performed for each user, but the number of users is more than that. It may be present or less. In addition, a plurality of users may share one user ID in order to form and use one person's personality in a department in a company, or one user may have a plurality of user IDs. It is also possible to.

The user terminal information 24a2 is unique information of the user terminal 10 for identifying each of the user terminals 10. For example, information is correctly distributed from the search server 20 such as the MAC address, serial number, and device ID to the user terminal 10. This is the information needed for this. Further, the information 24a3 required for controlling the distribution information to the user terminal 10 is a plurality of parameters set according to the taste and usage of each user such as the distribution frequency and the number of distributions.

Further, the technical interest information 24a4 is information related to the user's interest, which is set by the user prior to searching for information related to the technology via the user terminal 10. The technical interest information 24a4 is, for example, a keyword related to a technology or a technical field in which the user is interested, and is a keyword input by the user from a user terminal 10 such as cancer, pharmacy, AI, robotics, or a sensor, or a search. A keyword or the like selected by the user from the keywords provided by the server 20 is set.

The information 24a1 to 24a4 stored in these user databases 24 described above are registered in the search server 20 using the user terminal 10 when the user uses the search service using the search system 1 according to the present embodiment. It becomes the information to be done. Since it is not necessary to retain personally identifiable information (for example, address, name, etc.) in the user database 24, anonymous registration and use are possible in the search system 1.

(B) Technical information database 24b
The technical information database 24b is a database that stores widely collected information on various technologies, and processes information downloaded from an external technical information database that can be connected to the search server 20 and the downloaded information for the search system 1. The information or the information about the technology directly input by the search server 20 is stored.

In this embodiment, as information on technology, academic papers and the like submitted to academic journals in various fields such as chemistry, engineering, pharmacy, medicine, and architecture, and technologies such as technical directions published by companies on websites and the like. It targets various technical information (literature information) such as information contained in documents, patent information, and technical manual documents.

All of these technical information can be searched regardless of the type of technical document by matching the types and contents of the items to be stored when the original technical information database is stored in the technical information database 24 according to the present embodiment. It can be included. Of course, the search system 1 may search only the technical information published in the academic journal, or may search only the other technical information.

FIG. 6 is a diagram showing a structure of document information stored in the technical information database 24b included in the storage unit 24 according to the present embodiment.
As shown in the figure, in the present embodiment, as technical information, technical information ID 24b1, title 24b2, technical information outline 24b3, technological development progress 24b4, technical tag 24b5, application tag 24b6, and research and development institution 24b7 Technical information such as the country name of the research and development institution 24b8, the year of publication 24b9, the information acquisition source information 24b10, other items 24b11, and the detailed URL 24b12 are stored for each technical information.

The technical information ID 24b1 is unique identification information assigned to all the document information to be searched. The title 24b2 is index information of the technical information indicated by, for example, the title of an academic paper or a technical report. The technical information summary 24b3 is information that briefly describes the technical content of the literature information, and is information corresponding to an abstract of a treatise. The technological development progress 24b4 is information on an index indicating the technological development progress related to the technology which is the subject of the literature information at the seven stages defined in FIG.

Here, FIG. 7 is a diagram for explaining an index that defines the progress of technological development regarding individual literature information used in the search system 1 according to the present embodiment in multiple stages.
In this embodiment, as an example, the progress of technological development is classified into seven Technology readiness levels (TRL) as shown in the figure.

For example, level 0 means a level of "an idea that has no theoretical support to realize a designed function", and is the level at which the progress of technological development is the lowest. On the other hand, level 6 means a level at which the product is sold, and is the level at which the progress of technological development is the highest. Then, between levels 0 and 6, five levels of levels 1 to 5 are defined. It should be noted that the level classification of the progress of technological development is not limited to these seven, and the level may be divided by the number of divisions higher or lower, and the definition of each level is also included in the definition described in the present embodiment. Not limited. By defining TRL, it is possible to grasp the degree of progress of the technology (whether it is in the research and development stage or the technology close to the product, etc.) that the user is interested in on the system.

Further, the technical tag 24b5 is a keyword of the technology for characterizing the literature information, and for example, words / words of the technology used in the technical paper such as sensors, radio waves, materials, and polymers are used. Also. The use tag 24b6 is a keyword in the market / technical field in which the technology set in the technical tag 24b5 is utilized, and for example, words / words such as automatic, robot, aircraft, and automobile are used.

As other information, the research and development institution 24b7 that researched and developed the technology described in the literature information or published the technical literature, the country name of the research and development institution to which the research and development institution belongs 24b8, and the publication of the literature information. Year 24b9, information on the type of information acquisition source (for example, papers, patents, product catalogs, etc.) of the literature information (information acquisition source information) 24b10, attributes of the technology such as author information and publication journal name related to the literature information, etc. The technical information database 24 stores information about the other item 24b11 of the above and the detailed URL 24b12 indicating the information of the website on which the document information is published. A plurality of detailed URLs 24b12 can be retained for each technical information ID, and even if one URL becomes unusable due to site relocation or closure, another URL can be accessed. It is maintained so that the search user can check the detailed information without any problem.

(C) Search matter information database 24c
FIG. 8 is a diagram showing a configuration of case information stored in the technical case information database 24c included in the storage unit 24 according to the present embodiment.
The search matter information database 24c according to the present embodiment is a database that stores information given for each user search unit in association with the user ID 24a1 of the user data stored in the user database 24a. As shown in FIG. 8, in the present embodiment, a search matter unit such as user ID 24c1, matter ID 24c2, matter classification 24c3, search keyword c4, search condition 24c5, search status 24c6, number of evaluations 24c7, number of evaluations 24c8, number of high evaluations 24c9. Stores information about. Here, the case category 24c3 is information related to the search mode described later, and in the present embodiment, two modes, a limited express mode and a fixed point observation mode, are prepared.

Note that the user ID 24c1 is the same as the user ID 24a1 shown in FIG. In addition, the specific information stored in the search condition 24c5 is other than the information input as a keyword when the user searches for a desired technology, such as the area information of the research institution, the year of technology announcement, and the progress of technology development. There are several pieces of information related to the technology. Further, the search status 24c6 indicates whether or not the user has completed the search.

As will be described later, the completion of the search can be decided by the user himself, or can be automatically completed by setting arbitrary conditions such as when the search process is performed a certain number of times. The completed matter information is retained for a certain period of time, and it is possible to retain what kind of evaluation was performed by each user under what kind of search conditions. Furthermore, the number of evaluations 24c7 is the number of times the search server 20 has delivered technical information to the user, and the number of evaluations 24c8 is the cumulative value obtained by the user evaluating the technical information delivered in the past from the search server 20. is there. For example, the embodiment shown in FIG. 8 shows a case where each user evaluates 10 technical information items per distribution.

The high evaluation number 24c9 is the number of cases that were highly evaluated by the user out of the evaluation number 24c8. That is, as shown in FIG. 8, in the case of the case AAAA of user ID 0001, a total of 5 distributions are performed, and the user evaluates 10 technical information items per distribution, for a total of 50 cases. It shows that 10 technical evaluations out of the technical information of were highly evaluated. For example, if this information is used to calculate the ratio of the number of evaluations 24c8 and 24c9, it is possible to grasp on the system how much the delivered technical information matches the interests of the user.

In addition, if the search status 24c6 is completed with a small number of evaluations 24c7 and a high ratio of the number of evaluations 24c8 and 24c9, it is possible to recommend more efficient technical information on the system. Can be grasped. By adjusting the relevance score calculation method and parameters described later according to the user's case-based search results grasped on the system, it is possible to search for more optimal technical information for each user. it can.

(D) Technical information list database 24d
FIG. 9 is a diagram showing a configuration of case information stored in the technical information list database 24d included in the storage unit 24 according to the present embodiment.
As shown in the figure, the technical information list database 24d stores the search results of the document information performed by the search server 20 based on the keyword or the like set by the user in association with the user ID 24d1 and the matter ID 24d2. The user ID 24d1 and the matter ID 24d2 are the same as the user ID 24c1 and the matter ID 24c2 shown in FIG.

Here, as the search result of the document information stored in the technical information list database 24d, the corresponding technical information ID 24d3 is extracted from the technical information stored in the technical information database 24b according to the search keyword and the search condition associated with the matter ID 24d2. And store it. The number of searches can be set to any number.

In addition, as other information stored in the technical information list database 24d, various types such as clustering result 24d4, other calculation result 24d5, and relevance score 24d6, which are calculated by performing calculations as described later on the extracted technical information. The calculation result can be mentioned. Here, the clustering result 24d4 is a cluster based on the distance calculated based on the clustering method adopted in each project. As the clustering method, a general clustering method such as the k-means method, which is a non-hierarchical method, the shortest distance method, the longest distance method, the group averaging method, and the Ward method, which are hierarchical methods, can be adopted.

As mentioned above, it is possible to determine what kind of search result was obtained when which clustering method was adopted from the information stored in the technical information matter database. In addition, the other calculation result 24d5 can hold the calculation result by machine learning other than the clustering method. For example, as a calculation method, general methods such as neural networks and Bayesian inference can be mentioned. With this configuration, the technical information list database 24d and the search matter information list database 24c are linked by the user ID and the matter ID, so which calculation method was effective in which matter of which user and which matter was not effective. It can be grasped on the system.

(E) User evaluation log database 24e
FIG. 10 is a diagram showing a configuration of user evaluation information stored in the user evaluation log database 24e included in the storage unit 24 according to the present embodiment.
As shown in the figure, the user evaluation log database 24e has an evaluation result 24e5 such as "interested" or "not interested" by the user via the user terminal 10 for each technical information presented by the search server 20. The information related to the evaluation is stored in association with the user ID 24e1, the matter ID 24e2, and the technical information ID 24e3. In the present embodiment, as the evaluation information stored together with the evaluation result 24e5, the evaluation number (evaluation NO24e4), the result of whether or not the user has confirmed the detailed information of the technology at the time of evaluation (detailed confirmation presence / absence 24e6), and the evaluation Information such as the time 24e7, other evaluation parameters 24e8, and whether or not the bookmark was made (bookmark 24e9) is used.

<About the relevance score 24d6>
Next, the relevance score 24d6 stored in the technical information list database 24d shown in FIG. 9 will be described in detail. In the present embodiment, the relevance score 24d6 of the technical information indicates a numerical index expressing to what extent a user is interested in the technical information in a certain project. The relevance score 24d6 is calculated by the relevance score calculation unit 27 of the search server 20.

The relevance score calculation unit 27 calculates the relevance score 24d6 by using the data stored in the technical information database 24b, the technical information list database 24d, and the user evaluation log database 24e in the storage unit 24. Specifically, in the relevance score calculation unit 27, the technical information evaluated by the user as "interested" is the technical information directly recognized by the user as highly evaluated, and therefore the technical information evaluated by the user as "not interested". A relatively high relevance score of 24d6 is calculated in comparison with (low-rated technical information) and unrated technical information.

In the present embodiment, as will be described later, the relevance score 246d of the unrated technical information is calculated based on the similarity with the evaluated technical information. Therefore, as the relevance score 24d6 of the highly evaluated technical information, a value that is extremely high compared to the total value of several to several tens of similarities, for example, a value such as 999 or 9999 is calculated as the relevance score 24d6. To do. On the other hand, the relevance score calculation unit 27 compares the technical information evaluated by the user as "not interested" with the highly evaluated technical information and the unrated technical information because the technical information is directly recognized as a low evaluation by the user. Therefore, a relevance score of 24d6 having a relatively low value is calculated.

In the present embodiment, the relevance score 24d6 of the low-rated technical information is calculated, and a value such as -999 or -9999 is calculated as the relevance score 24d6. The calculation method of the relevance score 24d6 is not limited to the above, and the relevance score of the highly evaluated technical information is relatively higher than the relevance score of the unevaluated technical information, and the low evaluation technical information. It is possible to set the relevance score of unevaluated technical information to be relatively lower than the relevance score of unevaluated technical information.

In the present embodiment, the relevance score 246d of the unrated technical information is calculated based on the similarity with the evaluated technical information. Here, a general vector space document comparison method (for example, cosine similarity, Word2Vec, Retrofitting, etc.) can be used to calculate the similarity between technical information. For example, in the calculation of cosine similarity, each technical information is represented by a vector based on the word appearance frequency of text information such as the title of technical information and the outline of technical information, and as a result of calculating two vector similarity, " It is determined that the closer to "1", the higher the similarity between the technical information, and the closer to "0", the lower the similarity between the documents. The calculation of the degree of similarity is not limited to the above text information, and may be calculated by utilizing the attributes of other technical information.

The relevance score calculation unit 27 first calculates the similarity between the plurality of technical information evaluated by the user and the unrated technical information by using the document comparison method as described above. Next, the relevance score calculation unit 27 calculates the relevance score 24d6 from the calculated relevance based on a predetermined calculation logic.

In the present embodiment, for example, the method shown in FIG. 11 is used as the calculation logic for calculating the relevance score of the unrated technical information (technical information for which the score is calculated). FIG. 11 is a diagram for explaining an example of the calculation logic performed by the relevance score calculation unit 27 according to the present embodiment. In the present embodiment, the user is presented with four technical information, and each of the information is evaluated as "interested" or "not interested", and the four technical information 01 to 04 evaluated are not yet evaluated. The similarity of each of the technical information 05 of the evaluation is calculated, and the similarity is subjected to a predetermined process to calculate the relevance score.

In the present embodiment, the similarity between the technical information evaluated as interesting (highly evaluated technical information) and the unrated technical information (technical information for which the score is calculated) is the value as it is, and the score for calculating the relevance. Used as. On the other hand, the degree of similarity between the technical information evaluated as uninterested (low-rated technical information) and the technical information to be scored is set with a predetermined threshold value, and the degree of relevance is based on the comparison result with the threshold value. Calculate the score for calculation. For example, if the similarity is equal to or higher than the threshold value, the similarity with the low-rated technical information is high, so that the relevance score calculation unit 27 calculates a lower score. On the other hand, if the similarity is less than the threshold value, the similarity with the low-rated technical information is low, so the relevance score calculation unit 27 calculates a score that is not considered in the relevance score calculation.

The calculation of the relevance score will be explained in detail below. In the example shown in FIG. 11, the user evaluates the technical information 01 and 02 as interested, and the user evaluates the technical information 03 and 04 as not interested. First, the relevance score calculation unit 27 recognizes the technical documents 01 and 02 evaluated as interesting as highly evaluated technical information, and the highly evaluated technical documents 01 and 02 and the unevaluated technical documents (score calculation targets). After calculating the similarity of (technical document) 05, the similarity is used as the adjusted score without adjusting the value. That is, when the degree of similarity between the technical information 01 and the technical information 05 is calculated to be 0.453, the value of 0.453 is calculated as the adjusted score without adjusting the value. Similarly, the relevance score calculation unit 27 calculates the similarity with the technical information 05 for the technical information 02 evaluated as being interested, and if the value is 0.023, the value is not adjusted without adjusting the value. The value of 0.023 as it is is calculated as the adjusted score.

In the present embodiment, a threshold value is set for the similarity between the low-rated technical information and the technical information for which the score is calculated, and the score for calculating the relevance is calculated based on the comparison result with the threshold value. In the following, the adjusted score is calculated with the threshold value set to 0.5, but the threshold value may be other than that, and a plurality of threshold values may be provided instead of one. Good. In the present embodiment, the relevance score calculation unit 27 calculates the similarity between the low-rated technical information 03 and the technical information 05 for which the score is calculated, and if the value is 0.567, the threshold value is 0.5. Therefore, -0.1 is calculated as the adjusted score. The relevance score calculation unit 27 similarly calculates the similarity with the technical information 05 for the low-rated technical information 04, and if the value is 0.201, the threshold value is less than 0.5, so the adjusted score. Is calculated as 0. Then, as the relevance score of the unrated technical information 05 to be evaluated, the total value of the four adjusted scores adjusted for the similarity with the above-mentioned technical information 01 to 04 is given as the relevance score of the technical information 05. To do.

As described above, in the present embodiment, the evaluation result by the user is used for the calculation of the relevance score 24d6, such as lowering the value of the similarity calculated when the similarity exceeds a certain threshold value and adding them together. We are trying to reflect it more. As a result, as the evaluation is repeated, the technical information displayed on the user terminal 10 comes closer to the technical information having the content that suits the user's preference, so that the user can perform the evaluation action more interestingly. Can be given.

In calculating the relevance score 24d6, in addition to the evaluation result 24e5 stored in the user evaluation log database 24e, information such as detailed confirmation presence / absence 24e6, evaluation time 24e7, other evaluation parameters 24e8 or bookmark 24e9, etc. May be used as ancillary or as the main information for evaluation. Further, the calculation logic for calculating the relevance score is not limited to the above method, and other methods may be used. It is also possible to prepare a plurality of calculation logic patterns, and it is also possible to further improve the accuracy by selecting an arbitrary calculation logic according to the search scene of technical information, the evaluation result of the user, and the like.

In this way, the relevance score calculation unit 27 calculates the relevance score 24d6 for all the technical information stored in the technical information list database 24d and stores it in the technical information list database 24d.

Next, the operation of the search system 1 according to the present embodiment will be described.
<Setting of user's technical interest>
FIG. 12 is a flowchart for explaining a method of setting the user's technical interest information 24a4 in the user terminal 10 of the search system 1 according to the embodiment of the present invention. Further, FIG. 13 is a flowchart for explaining a method of setting the user's technical interest information 24a4 in the search server 20 of the search system 1 according to the embodiment of the present invention. Further, FIG. 14 is a sequence diagram showing an example of a procedure for setting technical interest information in the search system 1 according to the embodiment of the present invention.

In the present embodiment, the user who intends to search for technical information using the user terminal 10 sets a keyword related to the technology in which the user himself / herself is interested before searching for the technical information. In the present embodiment, the user selects the function list "interest setting function" displayed on the screen of the user terminal 10 to start the input process of the technical interest information (YES in step S101). As a result, the user terminal 10 requests the search server 20 for the keyword (steps S102, A1).

The search server 20 accepts a keyword request from the user terminal 10 (YES in step S201). In response to the keyword request from the user terminal 10, the search server 20 utilizes the data and the like stored in the technical information database 24b and the technical information list database 24d to extract keywords to be selected by the user (step S202). ). The keyword extraction in the search server 20 is performed, for example, from the technical information stored in the technical information database 24b, the technical tag of the technical information having a large number of highly evaluated (selected as being interested) through the evaluation by various users in the past. The keywords stored as 24b5 and the keywords stored as usage tags 24b6 are counted, and the top frequently used keywords are extracted.

Further, as a simpler method, keywords that are generally considered to be of high interest, such as general technical keywords that have been set in advance, may be estimated from database information and extracted as candidates. Further, the keyword may be extracted based on the text information of the technical information summary 24b3 stored in the technical information database 24b.

The search server 20 transmits the extracted plurality of keywords to the user terminal 10 (steps S203, A2). The user terminal 10 receives a keyword from the search server 20 (step S103), and displays the received plurality of keywords on the screen (step S104). Next, the user selects one or more keywords of interest from the plurality of keywords displayed on the screen of the user terminal 10, and selects either "keyword update" or "setting completed" after the keyword selection is completed. (Step S105). As a result, when "keyword update" is selected in step S105, the user terminal 10 transmits information regarding the keyword update to the search server 20 and requests the keyword update (steps S106, A3).

When the search server 20 receives the keyword update information and the keyword information transmitted from the user terminal 10 (in the case of keyword update in step S204), the technical information database 24b is searched with the selected keyword and added to the corresponding technical ID. Another keyword is extracted from the technical ID, and is transmitted to the user terminal 10 again (steps S205 and A4). The user terminal 10 returns to the process of receiving the keyword from the search server 20 (step S103), and repeats the above process until "setting complete" is selected in the keyword selection process (step S105) by the user. On the other hand, when "setting completed" is selected in step S105, the user terminal 10 transmits the selected keyword to the search server 20 (steps S107, A5). The search server 20 receives the information and the keyword regarding the setting completion transmitted from the user terminal 10 (when the setting is completed in step S204), and associates the received keyword with the user ID 24a1 of the user who uses the user terminal 10 that has transmitted the received keyword. It is stored in the user database 24a as technical interest information 24a4 (step S206).

<Search process for information related to technology>
Hereinafter, a document search process using the search system according to the embodiment of the present invention described above will be described.

FIG. 15 is a flowchart for explaining a search process for technical information in the user terminal 10 of the search system 1 according to the embodiment of the present invention. Further, FIG. 16 is a flowchart for explaining a search process for technical information in the search server 20 of the search system 1 according to the embodiment of the present invention. Further, FIG. 17 is a sequence diagram showing an example of a document search processing procedure in the search system 1 according to the embodiment of the present invention. 18 to 23 show an example of a screen displayed on the user terminal 10 at each stage of the technical information retrieval process according to the present embodiment.

The user operates the user terminal 10 to select the technical information search mode (step S301). FIG. 18 is a diagram showing an example of a mode selection screen displayed on the user terminal 10 at the start of the document search process in the search system 1 according to the embodiment of the present invention. Here, in the "fixed-point observation mode", the work of periodically checking the technical information within the range of interest of the user for a relatively long period of time (for example, several months) is performed, and any arbitrary location is provided. This mode is selected when it is desired to obtain a list of technical information (for example, a reference list) according to the purpose at the timing. Further, the "express mode" is a mode selected when it is desired to perform a search process to obtain technical information such as a desired document list in a relatively short time. A service provider that provides technical information such as a reference list that has received search results based on the limited express mode automatically obtains part or all of the technical information related to the search results stored in the search server 20 based on the search results. It may be transmitted to the user terminal 10, or the appropriate technical information selected by the service provider may be provided to the user based on the search result, for example, over a period of one to several days. You may.

Here, in the "first extraction unit" and "second extraction unit" described in claim 7, the user has selected either "fixed point observation mode" or "express mode". In this case, it means that the control unit 21 extracts information about a predetermined technique according to the purpose from the storage unit 24 and the communication unit 25 transmits it to the user terminal 10.

In the case of processing in the limited express mode, the user may input a keyword and use the keyword search result together with the extraction of technical information. In the present embodiment, the above two modes are prepared so that they can be selected by clicking the fixed point observation mode button 1801 or the limited express mode button 1802 on the screen, respectively.

Here, when the user selects the "fixed point observation mode" 1801 on the user terminal 10 (in the case of the fixed point observation mode in step S302), the user terminal 10 includes at least one of the user ID and the user terminal information. , The predetermined information such as the information that the fixed point observation mode is selected is transmitted to the search server 20 (steps S303 and B1). The purpose of the user to select the "fixed point observation mode" is, for example, to regularly check the technical information related to his / her interest target and save the technical information of interest.

The search server 20 is triggered by the fact that the information regarding the fixed point observation mode transmitted from the user terminal 10 is received via the communication unit 25 (YES in step S401), and the received user ID or user terminal information or the like is predetermined. Based on the information, the technical interest information 24a4 of the user is extracted from the information stored in the user database 24a in the storage unit 24, and the search condition for the technical information is created (step S402). For example, when the user database 24a is set to be interested in "digital healthcare" and "diabetes" as the technical interest information of the user, the control unit 21 uses these technical information databases 24b in the storage unit 24. Search by the keyword or related keyword, store the information about the corresponding technical information in the technical information list database 24d, and add a new search matter information list record to the search matter information database 24c (step S403).

The number of technical information stored in the technical information list database 24 is not particularly limited, and for example, thousands to tens of thousands of technical information may be listed. In consideration of implementation, the number of technical information that can be listed may be limited. This process is not limited to the first time, and may be performed every time the user selects the fixed point observation mode, and the technical information list database 24d may be updated each time. Further, even if the user is interested in a plurality of technologies at the same time, a plurality of search methods may be set / adjusted. In the fixed point observation mode, for example, the user's interest in technology is evaluated for several months, and the user's interest may change in the middle. Therefore, the same technical information is evaluated again. It may be included. In this way, the user can acquire the desired technical information even when the user's interest in the technology changes.

Then, the search server 20 extracts the technical information to be evaluated by the user and transmits it to the user terminal 10 (steps S404 and B2). In the present embodiment, the search server 20 extracts four technical information and sends it to the user terminal 10 so that the user can evaluate it. The extraction target of the technical information is determined based on the values of the clustering result 24d4 and the other calculation results 24d5 in the technical information list database 24d so that a certain amount of the technical information evaluated by the user is included. For example, an example is a determination method in which technical information that is as close to the center of gravity of different clusters as possible, that is, representative of the cluster, is to be extracted. From the second time onward, the relevance score 24d6 is calculated for each technical information ID unit by the user's evaluation, so the technical information that is more likely to be evaluated by the user as being more interested is extracted from the magnitude of the relevance score. It becomes possible.

On the other hand, when the user selects the "express mode" on the user terminal 10, the screen transitions to the search condition setting screen as shown in FIG. 19 (step S304). FIG. 19 is a diagram showing an example of a search information setting screen in the limited express mode displayed on the user terminal 10 according to the present embodiment. For example, the user can input predetermined data by touching each item portion of the search information setting screen of FIG. 19 displayed on the user terminal 10. In the present embodiment, the user uses each item button displayed on the user terminal 10 (for example, search keyword 1901, publication year (From) to publication year (To) 1902, research organization division 1903, regional condition 1904, technological development. Touch the degree of progress (1905, etc.) with your finger or the like, enter the search conditions for each item, and touch the search button 1906. By this action, the user terminal 10 sends the information regarding the search condition of the input technical information to the search server 20 together with at least one of the user ID and the user terminal information and the information that the limited express mode is selected. Transmit (steps S305, B1). One example of the purpose for the user to select the "express mode" is that the target of the specific technical information to be searched is clear and the user wants to obtain the search result matching the target more quickly.

The search server 20 is triggered by receiving the above information transmitted from the user terminal 10 via the communication unit 25 (step S401), and the control unit 21 receives the received user ID or user terminal information, and the user terminal 10. Based on the information regarding the search condition of the technical information input in, the following search process is executed to add the record to the search matter information list database 24c.

That is, the search server 20 creates a search condition for technical information based on the information of the search condition set on the user terminal (step S402). For example, search the technical information database 24b with detailed search conditions such as "digital healthcare", "announced from 2008 to 2018", "research by private companies", "no area limitation", etc. as keywords, and search for each of them. A set of technical information corresponding to is stored in the technical information list database 24d, and a new record is added to the search matter information database 24c (step S403). The set of technical information to be searched may be a large number of data such as thousands to tens of thousands.

Next, as in the "fixed point observation mode", the document list generation unit 26 of the search server 20 receives a plurality of technical information (for example,) to be evaluated by the user from the stored information of the technical information list database 24d of the storage unit 24. Extract 4 technical information).

Then, the search server 20 transmits the extracted technical information to the user terminal 10 as technical information for the user to evaluate for the first time (steps S404, B2). In the present embodiment, the number of technical information to be evaluated by the user is 4, but the number of technical information to be evaluated by the user at one time is not limited to, for example, 10 or 30, and the administrator of the search system 1 or the like can use it. It may be set freely.

When the user terminal 10 receives four technical information from the search server 20 (YES in step S306), the user operates the operation unit 12 of the user terminal 10 to evaluate the four received technical information. Is performed, and the evaluation result is transmitted from the user terminal 10 to the search server 20 (steps S307 and B3). The evaluation result may be transmitted when the evaluation work is completed, or may be transmitted together with the filter setting information set through the filter setting screen described later.

In addition, the evaluation work referred to here is the work of selecting the technical information that the user wants, such as whether the technical information the user wants to search or whether the technical information the user wants to know more details. Is. FIG. 20 is a diagram showing an example of a technical evaluation screen displayed on the display unit 13 of the user terminal 10 when the user performs the evaluation work. In the present embodiment, the user browses the technical information displayed on the display unit 13 of the user terminal 10 and evaluates whether or not he / she is interested in the technical information. In this embodiment, as shown in FIG. 20, technical information 2001 such as a technical title, a technical outline such as a treatise abstract, a development institution name, a development base, a developer name, and a country name is displayed on the display unit 13 for each case. If the user is interested in each of the technical information, he / she selects the "interested" button 2002, and if he / she is not interested, he / she selects the "not interested" button 2003 to evaluate the technical information. The manifestation of "interested" or "not interested" may be selected by any method such as voice input or swipe processing as well as button selection. As described above, in the present embodiment, the user can evaluate the technical information interactively and with a simple operation. Then, the technical information in which "interested" is selected by the user is regarded as the technical information highly evaluated by the user, and the technical information in which "not interested" is selected is regarded as the technical information in which the user is evaluated low.

As items displayed on the display unit 13 of the user terminal 10, in addition to the above items stored in the technical information database 24b, evaluations of other users calculated from the data stored in the user evaluation log database 24e The result etc. may be included. In addition, if it is difficult to evaluate because the displayed technical information is not enough, for example, by tapping the screen, the Internet web page that is the source of the information can be displayed and detailed information can be confirmed. .. As described above, the user determines "interested" / "not interested" for each technical information based on the displayed information or the like, but the user terminal 10 determines the result regarding the presence or absence of interest determined by the user. Not only that, other information related to the evaluation such as the time taken for the evaluation until the user made the decision and whether or not the detailed information has been confirmed may be retained as parameters.

Further, in the above evaluation work (step S307), the user can bookmark the technical information of interest in addition to the selection of "interested" / "not interested" in the technical information. In the present embodiment, the bookmark button 2004 is included in the screen displayed on the user terminal 10. With this bookmark function, it is possible to retain the technical information that is "not interested" at the time of evaluation, but that you want to make a note of as it may be used at a later date. By selecting the bookmark function separately for whether or not it is interested in this way, it is possible to eliminate the situation where the technical information selected as "interested" includes the technical information "simply want to make a note". It can be done and has the effect of improving the evaluation accuracy.

The user terminal 10 transmits the evaluation result to the search server 20 in the first evaluation work (step S307), but also includes evaluation parameters such as the time required for evaluation and filter condition setting information described later. It may be included and transmitted to the search server 20. Further, the evaluation result may be transmitted to the search server 20 when the evaluation of the four technical information is completed, or may be transmitted from the user terminal 10 to the search server 20 for each technical information. ..

When the user performs the above-mentioned evaluation work (step S307) on the four technical information transmitted from the search server 20 on the user terminal 10, the display of the display unit 13 of the user terminal 10 is as shown in FIG. Filter setting screen is displayed (step S308). FIG. 21 is a diagram showing an example of a filter setting screen displayed on the display unit 13 of the user terminal 10. On the filter setting screen, as shown in FIG. 21, typical keywords (for example, bone, diabetes, etc.) of technical information evaluated by the user in the evaluation work (step S307) and other attributes (for example, old technology, university) are displayed. Information 2101 regarding (departure, etc.) is displayed. When the user selects a keyword or the like that is not considered when selecting technical information from these keywords or the like displayed on the display unit 13 and selects the execute button 2102, a filter condition (exclusion condition) is set and the filter is concerned. The setting information is transmitted from the user terminal 10 to the search server 20 (steps S309 and B4). The information selected from the information 2101 such as keywords can be identified by the user by changing the display mode such as the color and pattern of the button area. By selecting a keyword or the like on the filter setting screen, the effect of lowering the priority of the technical information included in the selected title or summary text and the technical information corresponding to the set attribute can be obtained.

The contents of the set filter conditions can be changed by selecting the search condition setting menu 1803 on the screen shown in FIG. 18 displayed when the search mode of the user terminal 10 is selected (step S301). You can move to the screen and check it there. FIG. 22 is a diagram showing an example of a search condition setting change screen displayed on the display unit 13 of the user terminal 10.

Further, after setting the keywords to be excluded on the filter setting screen, the control unit 11 of the user terminal 10 selects whether to evaluate the technical information again to the user or output the search results before that as a list. A screen as shown is displayed, and the user selects whether to continue the same evaluation (step S310). FIG. 23 is a diagram showing an example of a screen for prompting a user to input an instruction displayed on the display unit 13 of the user terminal 10.

On the other hand, when the search server 20 receives the first evaluation result or the like from the user terminal 10 (YES in step S405), the technical information list generation unit 26 searches for each technical information for each user based on the evaluation result or the like. The relevance score 24d6 for each case is calculated, the content of the technical information stored in the technical information list database 24d is updated, and the priority of the technical information list is updated (step S406). An example of the calculation method of the relevance score 24d6 is as described above using FIG. 11 and the like, and all the technical information stored in the technical information list database 24d by the same method is used for each user ID and search item. Calculate and update the relevance score.

The calculation method of the relevance score is not limited to the above rule, and in addition, the attribute value of the technical information list database 24d is used as the input value, and the evaluation parameter (evaluation result, bookmark result, evaluation) on the user terminal is used as the output value. It may be replaced by a general machine learning method that utilizes the time taken).

When the user selects the "evaluate again" menu 2301 on the display screen shown in FIG. 23 (evaluation again in step S310), the user terminal 10 transmits information to that effect to the search server 20 (step S311, B5). When the search server 20 receives the information (re-evaluation in step S407), the search server 20 sorts the technical information stored in the technical information list database 24d based on the score of the assigned degree of relevance, and then informs the user. Four technical information to be evaluated are extracted and transmitted to the user terminal 10 (steps S408 and B6).

In the present embodiment, as an example, the top four unrated cases are extracted as the next technical information to be evaluated and transmitted to the user terminal 10, but the other four cases are extracted. May be good. For example, the technical information to be evaluated is extracted not only from the technical information with the highest score but also randomly, or the upper score, the middle score, and the lower score are combined and extracted at a certain ratio. The system administrator may be able to set or switch according to the purpose of the search system 1, the usage environment, and the like. By performing the evaluation repeatedly in this way, the technical information displayed on the user terminal 10 comes closer to the technical information having the content that suits the user's preference, which is an experience that allows the user to perform an evaluation action more interestingly. Can be given to the user.

On the other hand, when the user selects the "output technical information list" menu 2302 on the display screen shown in FIG. 23 (output the technical information list in step S310), the information about that is transmitted from the user terminal 10 to the search server 20. It is transmitted (step S312, B7). When the search server 20 receives the information (outputs the technical information list in step S407), the search server 20 sorts the technical information stored in the technical information list database 24d based on the assigned relevance score, and the user A predetermined number of technical information provided to the user terminal 10 is extracted and transmitted to the user terminal 10 (steps S409 and B8). Then, the user terminal 10 receives the list of technical information from the search server 20 (step S313), and the user can browse the document information. In the present embodiment, 100 technical information items are extracted and output in order from the highest score, but an arbitrary number (for example, 50 items, 1000 items, etc.) of technical information is extracted by any other method. You may try to do the same.

FIG. 24 is a flowchart for explaining the details of the technical information list output process (step S409) of the search server 20 in the present embodiment. Here, if the search condition selected in step S301 is the "express mode" (in the case of the limited express mode in step S409a), the search server 20 outputs the list as it is (step S409b). On the other hand, when the mode set by the user is the "fixed point observation mode" (in the case of the fixed point observation mode in step S409a), after the above-mentioned relevance score assigning process, the search server 20 is ready to output to the user terminal 10. Notify that (step S409c). Then, the user terminal 10 sets the conditions of the technical information to be output from the search screen as shown in FIG. 16, and transmits the setting information to the search server 20 (step S310). The search server 20 extracts the technical information list according to the extraction conditions received from the user terminal 10 (step S409d), and outputs the extracted list to the user terminal 10 (step S409e).

In the present embodiment, the search application software according to the present invention is installed on the user terminal 10 and the search server 20. In the search application software, a predetermined process is performed on the search server 20 based on a user operation on the user terminal 10.

As described above, using the search system, search method, and search application software according to the present embodiment, the user repeatedly evaluates the technical information, calculates the similarity between the technical information, and extracts the technical information to be evaluated. By doing so, it becomes possible to appropriately extract technical information according to the interests of individual users from a vast amount of diverse technical information and provide it to users. In particular, the information retrieval mode is divided according to the purpose of the user's technical information retrieval (in this embodiment, two modes, a fixed point observation mode and a limited express mode), and the user is repeatedly evaluated within that range. It is to be able to output search results that suit your interests.

In addition, the search for technical information described in this embodiment does not utilize full-text search results of texts, but utilizes limited standard data in the same database. Therefore, papers, patent documents, and product information. , Researcher / engineer information, various technical information such as technical ideas can be applied.

Furthermore, it is also a feature that it can be applied regardless of the technical field. Depending on the technical field, it may often be difficult to judge the similarity between technical information due to the fact that the same word is used with different meanings or the rules described in the literature are different. , The above problem is solved by setting the user's evaluation result (the result judged by a person who seeks technical information) in the objective function. That is, even technical information that is difficult to search only with text information can be searched / extracted from the database based on the evaluation results of interest / non-interest judged by the user at the time of the search. Search results can be obtained according to the user's interests.

1 Search system 10 User terminal 20 Search server 30

Network

11, 21

Control unit

12, 22

Operation unit

13, 23

Display unit

14, 24 Storage unit 24a User database 24b Technical information database 24c Technical matter information database 24d Technical information list database 24e User

Evaluation log database

15, 25 Communication department 26 Technical information list generation department 27 Relevance score calculation department

Claims

A search system including a user terminal for a user to input search conditions and a search server that connects to the user terminal via a network and searches for information based on the search conditions.
The search server includes a technical information database that stores information on a plurality of technologies.
A first technical information extraction unit that extracts information on a plurality of first technologies to be evaluated by the user terminal from the technical information database, and
Based on the evaluation result of the information on the plurality of first technologies in the user terminal, the degree of similarity between the information on the plurality of first technologies and the information on the second technology stored in the technical information database is determined. The relevance score calculation unit that calculates the relevance score processed in
A technical information list generation unit that generates a list in which information about the plurality of technologies stored in the technical information database is sorted in descending order of the relevance score for each user.
A technical information list database that stores the list generated by the technical information list generation unit, and
A second technical information extraction unit that extracts information on a predetermined technology from the technical information list database based on the search conditions, and a second technical information extraction unit.
A search system characterized by being equipped with.
The relevance score calculation unit calculates the similarity between the information related to the plurality of first techniques and the information related to the second technique, and evaluates the similarity with respect to each of the information related to the plurality of first techniques. The search system according to claim 1, wherein an adjusted score adjusted based on the result is calculated, and the calculated relevance score is calculated by adding up the calculated adjusted scores.
The relevance score calculation unit is a positive number that is at least several times or more the adjusted score of the information about the first technique and the information about the second technique that the user highly evaluated, or the user is low. It is characterized in that a negative number of at least several times or more of the adjusted score of the evaluated information on the first technique and the information on the second technique is calculated as the relevance score of the information on the first technique. The search system according to claim 2.
The relevance score calculation unit includes information on the first technique and information on the second technique in which the evaluation result is highly evaluated when the similarity is expressed by a numerical value in the range of -1 to +1. The adjusted score is calculated by multiplying or adding a coefficient of 1.0 or more according to the value of the similarity of the above, and the information on the first technique having a low evaluation result and the first technique. The search system according to claim 3, wherein the adjusted score is calculated by performing adjustment by multiplying or subtracting a coefficient of 0 or less according to the value of the similarity with the information relating to the technique of 2.
The relevance score calculation unit adjusts the magnitude of the similarity in multiple steps based on the magnitude of the similarity between the information regarding the first technique and the information regarding the second technique. The search system according to claim 4, wherein the adjustment degree score is calculated.
The search server includes a user database for storing information about a technology that the user is interested in as interest information for each user.
The first technical information extraction unit is characterized in that, based on the interest information stored in the user database, information on the plurality of first technologies for causing a user to evaluate from the technical information database is extracted. The search system according to claim 5.
The search server includes a first extraction unit that extracts information about a predetermined technology from the technical information list database that stores information about the technology evaluated within a predetermined period.
A second extraction unit for extracting information on a predetermined technology from the technical information list database that stores information on the last evaluated technology without considering past evaluations is provided, and either extraction unit is used. The search system according to claim 6, wherein it is possible to select to extract information about the technology.
The user terminal displays information about the plurality of first technologies extracted by the first technical information extraction unit one by one, causes the user to determine whether or not he / she is interested, and determines that the user is interested. The search according to claim 7, wherein the information regarding the first technology is highly evaluated information, and the information regarding the first technology determined by the user to be uninterested is low-evaluated information. system.
It is characterized in that a calculation method or parameter used for processing of the relevance score calculation unit and the technical information list generation unit can be selected based on the evaluation result regarding the information related to the first technology evaluated by the user terminal. The search system according to any one of claims 1 to 8.
A search method using a search server that connects a user terminal into which a user inputs search conditions to the user terminal via a network and searches for information based on the search conditions.
A first step of extracting information on a plurality of first technologies for evaluation by the user terminal from a technical information database that stores information on a plurality of technologies included in the search server.
Similarity between the information on the plurality of first technologies and the information on the second technology stored in the technical information database based on the evaluation results regarding the information on the plurality of first technologies evaluated on the user terminal. The second step of calculating the relevance score after performing a predetermined process each time,
A third step of sorting the information related to the plurality of technologies stored in the technical information database in descending order of the relevance score for each user and storing the information in the technical information list database together with the relevance score.
A fourth step of extracting information about a predetermined technology from the technical information list database based on the search conditions, and
A search method characterized by having.
Search application software that can be executed on a search server that searches for information based on the search conditions, which can communicate with the user terminal on which the user inputs search conditions.
A first function for extracting information on a plurality of first technologies for evaluation by the user terminal from a technical information database for storing information on a plurality of technologies included in the search server, and a first function.
Similarity between the information on the plurality of first technologies and the information on the second technology stored in the technical information database based on the evaluation results regarding the information on the plurality of first technologies evaluated on the user terminal. A second function that calculates the relevance score after performing a predetermined process each time,
A third function of sorting information on the plurality of technologies stored in the technical information database in descending order of the relevance score for each user and storing the information in the technical information list database together with the relevance score.
The fourth function of extracting information on a predetermined technology from the technical information list database based on the search condition is
Search application software that can be executed on the search server.
The second function calculates the similarity between each of the information regarding the plurality of first technologies and the information regarding the second technology, and evaluates the similarity with respect to each of the information regarding the first technology. The search application software according to claim 11, wherein the adjusted score adjusted based on the above is calculated, and the calculated adjusted scores are added up to calculate the relevance score.
The second function is a positive number that is at least several times or more the adjusted score of the information about the first technology and the information about the second technology that the user highly evaluated, or the user gives a low evaluation. The claim is characterized in that a negative number of at least several times or more of the adjusted score of the information regarding the first technique and the information regarding the second technique is calculated as the relevance score of the first information. The search application software according to 12.
The second function includes information on the first technique and information on the second technique in which the evaluation result is highly evaluated when the similarity is expressed by a numerical value in the range of -1 to +1. The adjusted score is calculated by multiplying or adding a coefficient of 1.0 or more according to the value of the similarity, and the information on the first technique having a low evaluation result and the second technique. 13. The search application software according to claim 13, wherein the adjusted score is calculated by performing an adjustment of multiplying or subtracting a coefficient of 0 or less according to the value of the similarity with the information relating to the technique.
The second function adjusts the magnitude of the similarity in multiple steps based on the magnitude of the similarity between the information regarding the first technique and the information regarding the second technique. The search application software according to claim 14, wherein the degree score is calculated.
A fifth function of storing interest information about the user's technology in the user database provided in the search server can be further executed.
The first function extracts information about the plurality of first technologies from the technical information database that stores information about a plurality of technologies included in the search server, based on the matters of interest stored in the user database. 15. The search application software according to claim 15.
A first mode for extracting information about a given technology from the technical information list database that stores information about the technology evaluated within a predetermined period, and information about the last evaluated technology without considering past evaluations. 16. The search application software according to claim 16, wherein a sixth function of selecting a second mode for extracting information about a predetermined technology from the technical information list database for storing information is further feasible.
The first function relates to the first technique in which the user terminal displays information about the plurality of first techniques one by one to cause the user to determine whether or not the user is interested, and the user determines that the user is interested. The search application software according to claim 17, wherein the information is regarded as highly evaluated information, and the information related to the first technology determined by the user to be uninterested is regarded as low evaluated information.
The claim is characterized in that a calculation method or parameter used for processing the second function and the third function can be selected based on the evaluation result regarding the information regarding the first technique evaluated by the user terminal. The search application software according to any one of 11 to 18.