Information processing system, control method for information processing system, and program

Abstract

To make it possible to provide more accurate detection results.SOLUTION: An information processing system (100) includes acquisition units (2132, 2141) for acquiring sentences (51, 61) that belong to a second knowledge domain (5) different from a first knowledge domain and that are related to words and phrases (41) that belong to the first knowledge domain (4), a search unit (2142) for searching for information that belongs to the second knowledge domain based on the sentences acquired by the acquisition units, and an output unit (215) for outputting search results obtained by the search unit.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an information processing system, a control method for an information processing system, and a program. [Background technology]

[0002] In recent years, biomimetics has become popular as a way to develop new materials inspired by the structure, functions, and production principles of living organisms. In biomimetics, biology literature is often searched to find organisms with the same functions as the material being developed. For example, when searching biology literature using engineering terms as a key, it is necessary to link information from the different fields of biology and engineering, since different fields use different terms or expressions even for the same function.

[0003] Non-Patent Document 1 discloses a keyword search system based on ontology engineering and Linked Data technology for searching biomimetics databases for information that can lead to ideas for new technologies. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Transactions of the Japanese Society for Artificial Intelligence, "Keyword Search Based on Bioinspired Engineering Ontology and Linked Data," 2016 Summary of the Invention [Problem to be solved by the invention]

[0005] There is room for improvement in conventional search systems in search technology that spans different domains. One aspect of the present invention can provide a search system that can search for information by connecting different knowledge domains. [Means for solving the problem]

[0006] In order to solve the above problem, an information processing system according to one embodiment of the present invention includes an acquisition unit that acquires sentences that belong to a second knowledge domain different from the first knowledge domain and that are related to terms that belong to the first knowledge domain, a search unit that searches for information that belongs to the second knowledge domain based on the sentences acquired by the acquisition unit, and an output unit that outputs the search results by the search unit.

[0007] In order to solve the above problem, a control method for an information processing system according to one embodiment of the present invention is a control method for an information processing system that searches for information in a second knowledge area using information in a first knowledge area as a key, and includes an acquisition step of acquiring sentences that belong to the second knowledge area different from the first knowledge area and are related to terms that belong to the first knowledge area, a search step of searching for information that belongs to the second knowledge area based on the sentences acquired in the acquisition step, and an output step of outputting the search results obtained in the search step. [Effects of the Invention]

[0008] According to one aspect of the present invention, it is possible to provide a search system that is capable of searching for information by linking different knowledge domains. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram showing the configuration of a search system according to a first embodiment of the present invention. [Figure 2] 2 is a block diagram showing an example of a configuration of a main part of the search system shown in FIG. 1. FIG. [Figure 3] FIG. 2 is a diagram showing an example of an input screen for a first knowledge domain and a second knowledge domain. [Figure 4] FIG. 10 is a diagram showing an example of a word / phrase input screen. [Figure 5] FIG. 10 is a diagram illustrating an example of data stored in a related sentence database. [Figure 6] FIG. 10 is a diagram showing an example of a related sentence selection screen. [Figure 7] FIG. 10 is a diagram showing an example of literature list data of a search result. [Figure 8] 2 is a diagram showing an example of the flow of a search process executed by a control unit of the search server shown in FIG. 1. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will now be described with reference to the accompanying drawings. A search system 100 (information processing system) according to this embodiment is used for searches across different knowledge domains.

[0011] Here, a knowledge domain refers to a range of knowledge that a person can acquire and accumulate in a specific field. A field refers to a range established for classifying information. Specifically, a field may be a range established based on, for example, academic discipline, industry, region, time period, book classification, etc. For example, the first knowledge domain 4 and the second knowledge domain 5 are knowledge domains that belong to different academic fields.

[0012] In the following, an example will be described in which the first knowledge domain 4 is the engineering domain and the second knowledge domain 5 is the biology domain, but this is not the only example, and the specific knowledge domains are not particularly limited. For example, the first knowledge domain 4 or the second knowledge domain 5 may be a knowledge domain belonging to academic fields such as physics, chemistry, biology, engineering, economics, geography, philosophy, law, astronomy, geology, anthropology, medicine, etc., or a knowledge domain belonging to industries such as the automotive industry, food industry, IT industry, travel industry, finance industry, construction industry, agriculture industry, and space industry. For example, the first knowledge domain 4 and the second knowledge domain 5 may be knowledge domains belonging to different eras, countries, regions, areas, etc.

[0013] As a preferred example, the first knowledge domain 4 may be a knowledge domain belonging to fashion studies, design studies, architecture, ergonomics, systems engineering, information engineering, informatics, chemistry, applied chemistry, nutrition, food science, etc., and the second knowledge domain 5 may be a knowledge domain belonging to medicine, physiology, biochemistry, ecology, genetics, cell biology, molecular biology, microbiology, bacteriology, enzymology, environmental science, oceanography, evolutionary biology, embryology, etc.

[0014] Different knowledge domains may use different terms or expressions. For example, for the same function of "repelling water," the term "water repellency" may be used in engineering, while the term "self-cleaning, anti-fouling" may be used in biology.

[0015] For example, snail shells have the function of "water repellency," but in biology, terms such as "self-purifying" and "fouling-resistant" are used instead of "water repellency." Therefore, when developing a water-repellent material, even if a developer searches biological literature using the term "water repellency" to get inspiration from organisms with water-repellent properties, there is a high possibility that meaningful information for the developer (e.g., information about the function possessed by snails) will not be extracted. Furthermore, even if information about snails is extracted, there is a possibility that a large amount of literature that is unnecessary for the developer (e.g., literature unrelated to water-repellent properties) will also be extracted.

[0016] As such, different knowledge domains use different terms or expressions even for the same function, making it difficult to extract appropriate documents. To improve the accuracy of searches across different knowledge domains, it is necessary to link information across different knowledge domains.

[0017] As will be described in more detail later, the search system 100 can perform highly accurate detection by using a related sentence database 221 (knowledge database) that associates words 41 belonging to a first knowledge domain 4 with related sentences 51 belonging to a second knowledge domain 5 that indicate at least one of the functions, characteristics, actions, phenomena, or states to which the words 41 are related.

[0018] <Search System 100> Fig. 1 is a diagram showing the configuration of a search system 100. As shown in Fig. 1, the search system 100 includes a user terminal 1, a search server 2, and a literature database 3 that stores literature and the like in each knowledge domain. The user terminal 1 and the search server 2 are connected to each other so that they can communicate with each other via the Internet 9 or the like.

[0019] The user terminal 1 is a terminal used by a user who wishes to perform a search across different knowledge domains.

[0020] The search server 2 is a server that searches for information belonging to a second knowledge domain 5 using terms 41 belonging to a first knowledge domain 4. The search server 2 has a related sentence database 221 in which related sentences 51 belonging to the second knowledge domain 5 are registered in association with terms 41 belonging to the first knowledge domain 4. Note that the related sentence database 221 does not have to be provided within the search server 2, but may be stored in an external storage or the like that is communicatively connected to the search server 2. Details of searches by the search server 2 will be described later.

[0021] The literature database 3 includes literature databases for each field, such as field A literature database 31, field B literature database 32, and field C literature database 33. For example, field A literature database 31 is a database that stores literature etc. belonging to biology, field B literature database 32 is a database that stores literature etc. belonging to zoology, and field C literature database 33 is a database that stores literature etc. belonging to botany. In addition to these, the literature database 3 includes multiple databases that store literature etc. belonging to various academic fields.

[0022] Each database in the literature database 3 stores documents, etc., in one area to be searched as the second knowledge area 5. Any object that can be searched collectively as one database is acceptable, and there are no restrictions based on the name of the field. The literature database 3 also includes, in addition to documents, web articles, text data generated by transcribing audio or video data, various documents, etc. Specifically, it may be, for example, a database that aggregates and publishes papers, articles, etc. in a specific academic field, either paid or free of charge, or a patent document database from each country. Alternatively, it may include non-public materials accumulated within a specific organization.

[0023] The literature database 3 is communicably connected to the search server 2 via the Internet 9 or the like. Furthermore, without being limited to this, the literature database 3 may be stored in a storage unit 22 of the search server 2, a storage unit (not shown) of the user terminal 1, or the like, which will be described later.

[0024] 2 is a block diagram showing an example of the main configuration of the search system 100. Below, an example will be described in which a word 41 "water repellency" belonging to "engineering" which is a first knowledge domain 4 is used to search for literature in "biology" which is a second knowledge domain 5. Note that the first knowledge domain 4, the second knowledge domain 5, and the word 41 are not limited to these, and are not particularly limited.

[0025] As shown in FIG. 2, the user terminal 1 includes a control unit 11 that controls the individual units of the user terminal 1, an input unit 12, a display unit 13, and a communication unit 14 for communicating with other devices.

[0026] The input unit 12 is, for example, a keyboard and a mouse for inputting characters, etc. The display unit 13 is, for example, a liquid crystal display, which displays various information. The display unit 13 and the input unit 12 may be integrally configured using a touch panel display, etc.

[0027] The user terminal 1 starts a browser to access a site provided by the search server 2, receives screen data for searching from the search server 2, and displays it on the display unit 13 of the user terminal 1. The user can perform a search by entering various information into input fields and the like of the screen data displayed on the display unit 13 and sending it to the search server 2.

[0028] The search server 2 includes a control unit 21 that controls all the components of the search server 2, and a storage unit 22 that stores various data used by the control unit 21.

[0029] The control unit 21 includes a knowledge domain acquisition unit 211, a phrase acquisition unit 212, a related sentence acquisition unit 213, a literature list acquisition unit 214, an output unit 215, and a communication control unit 216. The memory unit 22 includes a related sentence database 221, a keyword set memory unit 222, and a literature list data memory unit 223.

[0030] As described above, the related sentence database 221 is a database that associates a term 41 belonging to the first knowledge domain 4 with one or more related sentences 51 belonging to the second knowledge domain 5 that indicate at least one of the functions, characteristics, actions, phenomena, or states to which the term 41 is related.

[0031] The keyword set storage unit 222 stores a keyword set 61, which will be described later, and the document list data storage unit 223 stores document list data searched by the document search unit 2142, which will be described later.

[0032] (Knowledge Area Acquisition Department 211) The knowledge domain acquisition unit 211 acquires information on the first knowledge domain 4 and the second knowledge domain 5 transmitted from the user terminal 1 via the communication control unit 216 .

[0033] FIG. 3 shows an example of an input screen for the first knowledge domain 4 and the second knowledge domain 5 displayed on the display unit 13 of the user terminal 1. The input screen includes a first knowledge domain input field 4a and a second knowledge domain input field 5a. By clicking or hovering the mouse over the first knowledge domain input field 4a and the second knowledge domain input field 5a, respectively, a plurality of knowledge domains can be displayed in the first knowledge domain input field 4a and the second knowledge domain input field 5a. The user can input the desired knowledge domain into the first knowledge domain input field 4a and the second knowledge domain input field 5a by selecting one knowledge domain from the displayed plurality of knowledge domains. Alternatively, the user may directly input the desired knowledge domain, such as "engineering," into the first knowledge domain input field 4a and the second knowledge domain input field 5a.

[0034] The input information on the first knowledge domain 4 and the second knowledge domain 5 is transmitted to the search server 2 via the communication unit 14 and acquired by the knowledge domain acquisition unit 211 via the communication control unit 216. In another embodiment, the input of the first knowledge domain 4 by the user may be omitted. In this case, the knowledge domain acquisition unit 211 may determine the first knowledge domain 4 based on the phrase 41 input by the user.

[0035] The user may also input a "search purpose" here. Examples of search purposes include collecting known information, searching for examples of applying ideas beyond knowledge domains, obtaining ideas for applying phenomena in other fields to other fields, creating a knowledge graph that represents the connections between various knowledge in a graph structure, and creating a new database using extracted information. By inputting the search purpose, the related sentences 51 used in the search, and the elements or parts of speech contained in the related sentences 51 may be weighted. Specifically, for example, if the phrase "water repellent" is entered with the purpose of obtaining new ideas, processing may be performed to set the importance of verbs or predicate verbs (e.g., "repellent"). If the purpose is to search for specific cases, processing may be performed to set the importance of verbs and objects to the same. This configuration makes it possible to control the quality of extracted information according to the user's needs.

[0036] (Word acquisition unit 212) Returning to Fig. 2, the explanation will be made again. The term acquisition unit 212 acquires terms 41 belonging to the first knowledge domain 4 transmitted from the user terminal 1 via the communication control unit 216. In another embodiment, the term acquisition unit 212 may identify terms that are synonymous with or similar in meaning to the terms 41. In another embodiment, the term acquisition unit 212 may identify terms obtained by translating the terms 41 into a language different from the input language.

[0037] 4 is a diagram showing an example of an input screen for a term 41 displayed on the display unit 13 of the user terminal 1. The user inputs "water repellent" as a search key into the term input field 41a. Information on the term 41 input by the user is transmitted to the search server 2 via the communication unit 14 and acquired by the term acquisition unit 212 via the communication control unit 216. Note that the term input field 41a may be displayed on the same screen as the first knowledge domain input field 4a and the second knowledge domain input field 5a.

[0038] Furthermore, the phrase 41 may be at least one of a single word and a compound word that is a combination of multiple words. In another embodiment, multiple phrases 41 may be input. The phrase 41 may be one or multiple words, a compound word, or a combination thereof.

[0039] (Related sentence acquisition unit 213) 2, the related sentence acquisition unit 213 includes a related sentence database selection unit 2131 and a related sentence data acquisition unit 2132.

[0040] The related sentence database selection unit 2131 selects an appropriate database from the related sentence database 221 stored in advance in the storage unit 22.

[0041] The related sentence database 221 is a database that associates a phrase 41 belonging to a first knowledge domain 4 with a related sentence 51 belonging to a second knowledge domain 5 that indicates at least one of a function, a characteristic, an action, a phenomenon, or a state related to the phrase 41. The phrase 41 is associated with one or more related sentences 51.

[0042] For example, when the second knowledge domain 5 is biology, the word 41 "water repellency" belonging to the first knowledge domain 4 "engineering" is associated with a related sentence 51 belonging to the second knowledge domain 5 that indicates a function that an organism can achieve by "water repellency." Also, for example, "water repellency" is associated with a related sentence 51 belonging to the second knowledge domain 5 that indicates a "characteristic" that an organism possesses, an "action" of an organism, a "phenomenon" that can occur in an organism, a "state" of an organism, or a "state" of a specific part of an organism.

[0043] Specific examples of sentences that describe a function in biology include "human skin regulates body temperature through sweat" and "burdock seeds stick to animal fur." ​​Examples of sentences that describe a characteristic in biology include "a leopard's spots blend into the grassland scenery," "a shark's skin structure reduces resistance in water," and "kangaroos have pouches on their abdomens." Examples of sentences that describe an action in biology include "birds build nests to breed," and "kangaroos carry their young in pouches." From another perspective, sentences that describe behavior may also be used. Examples of sentences that describe a phenomenon in biology include "cherry blossoms bloom in Japan in the spring," and "water droplets roll on the surface of leaves." Examples of sentences that describe a state in biology include "a hummingbird remains stationary in mid-flight," and "some reptiles cease activity due to a drop in body temperature." Note that the above are merely examples, and sentences with the same meaning but different expressions, or sentences expressed in other languages, may also be included.

[0044] In this way, by using the related sentence database 221 that associates words 41 with related sentences 51 belonging to the second knowledge domain 5 that indicate the function, characteristic, action, phenomenon, or state related to the words 41, it becomes possible to perform a search to obtain ideas for applying biological phenomena, etc. to engineering. In other words, it becomes possible to perform a search aimed at finding the biological species, biological parts, biological structures (components), mechanisms, etc. that realize the function, characteristic, action, phenomenon, or state based on the function, characteristic, action, phenomenon, or state based on the material to be developed.

[0045] Furthermore, the phrase 41 may also be a phrase indicating at least one of a function, a characteristic, an action, a phenomenon, or a state in the first knowledge domain 4. By specifying a phrase indicating a function, a characteristic, an action, a phenomenon, or a state in the first knowledge domain 4 as a search key, it is possible to search for information related to the function, characteristic, action, phenomenon, or state related to the phrase 41 in the second knowledge domain 5.

[0046] Specifically, for example, terms that indicate functions or properties in engineering include "elasticity," "conductivity," "heat insulation," and "transparency." For example, terms that indicate actions in engineering include "cutting," "heating," and "drying." For example, terms that indicate phenomena in engineering include "corrosion," "penetration," "refraction," and "resonance." For example, terms that indicate states in engineering include "wear," "expansion," and "vibration."

[0047] The related sentence database 221 stores multiple databases with different combinations of the first knowledge domain 4 and the second knowledge domain 5. For example, database D1 is a database in which engineering is the first knowledge domain 4 and biology is the second knowledge domain 5, and is a database in which a term 41 belonging to engineering is associated with a related sentence 51 belonging to biology that indicates at least one of a function, a characteristic, an action, a phenomenon, or a state to which the term 41 is related. Database D2 is a database in which engineering is the first knowledge domain 4 and botany is the second knowledge domain 5, and is a database in which a term 41 belonging to engineering is associated with a related sentence 51 belonging to botany that indicates at least one of a function, a characteristic, an action, a phenomenon, or a state to which the term 41 is related. Database D3 is a database in which engineering is the first knowledge domain 4 and zoology is the second knowledge domain 5, and is a database in which a term 41 belonging to engineering is associated with a related sentence 51 belonging to zoology that indicates at least one of a function, a characteristic, an action, a phenomenon, or a state to which the term 41 is related. In addition to these, the related sentence database 221 also includes various databases relating to the first knowledge domain 4 and the second knowledge domain 5.

[0048] The related sentence database selection unit 2131 selects, from the related sentence database 221, the database D1 acquired by the knowledge domain acquisition unit 211, which is a database relating to "engineering" and "biology", for example.

[0049] The related sentence data acquisition unit 2132 (acquisition unit) acquires from the database D1 related sentences 51 belonging to the second knowledge domain 5 "biology" that indicate at least one of the functions, characteristics, actions, phenomena, or states associated with the phrase 41 "water repellency" belonging to the first knowledge domain 4 "engineering" acquired by the phrase acquisition unit 212.

[0050] 5 is a diagram illustrating an example of data stored in database D1 of related sentence database 221. As shown in Fig. 5, "engineering" is stored in first knowledge domain column 4b, and "biology" is stored in second knowledge domain column 5b. "Water repellency" is a word 41 stored in word column 41b belonging to the first knowledge domain.

[0051] The related sentence column 51b stores one or more related sentences 51 belonging to the second knowledge domain 5, which indicate at least one of the functions, characteristics, actions, phenomena, or states to which "water repellency" is related. The related sentences 51 are stored separated into sentence components such as a subject, predicate, and object. The related sentences 51 need only include at least a subject and a predicate, and may also include a predicate verb, an object, a complement, a noun, a pronoun, a verb, an adjective, an adverb, an auxiliary verb, an article, a preposition, a conjunction, an interjection, a noun phrase, an adjective phrase, an adverb phrase, and the like.

[0052] In the example of Figure 5, six related sentences 51 are stored, separated into sentence components such as subject, term, and object. Column 1 stores "plant leaves resist adhesion," column 2 stores "young leaf roll water droplet," column 3 stores "heteropterid bugs run over the water's surface," column 4 stores "pitcher plants trap insects and other small prey," column 5 stores "bacteria digest oil," and column 6 stores "hair repel water."

[0053] In the example of FIG. 5, the related sentences 51 are written in English, but they may be written in Japanese or another language. Related sentences 51 in different languages ​​may be mixed. Similarly, phrases 41 may be stored in a language other than Japanese. Another aspect may be a collection of multiple phrases that are synonymous with or have similar meanings to "water repellent."

[0054] The related sentence data acquisition unit 2132 acquires data of the six related sentences 51 from the related sentence database 221 and transmits it to the user terminal 1 via the communication control unit 216 together with the screen data.

[0055] The related sentence database 221 is not limited to any particular type of information as long as it associates the phrase 41 with the related sentence 51. For example, the related sentence database 221 may be a lookup table that defines the correspondence between multiple pieces of information, a mathematical model that mathematically associates multiple pieces of information, a trained model that generates a specific output for any input, or a combination of two or more of these.

[0056] 5 may be created, for example, by the following method: First, the control unit 21 executes a literature search in the knowledge domain of "biology" corresponding to the second knowledge domain 5 (specifically, the field A literature database 31) using the phrase 41 "water repellency" as a search key.

[0057] Next, the control unit 21 performs natural language processing on the extracted document and further extracts important words and phrases from the document. Important words and phrases are, for example, words and phrases that are present near the word 41 "water repellent" (including translated words and synonyms) and that appear frequently in the document. In another embodiment, words and phrases that meet any conditions input by the user may be determined to be important words and phrases. When determining importance based on the number of times words and phrases appear in a document, the number of times they appear may be calculated by treating words and phrases with similar meanings as one group.

[0058] Next, the control unit 21 combines the extracted important words and phrases to create one or more related sentences 51 that can be understood by humans or AI (Artificial Intelligence). Here, the related sentences 51 may be created by, for example, a method of detecting co-occurring words using a known text mining tool, a method of using a known natural language processing model, or the like. For example, the number of related sentences 51 created by the control unit 21 may be determined based on a user input. Alternatively, the related sentences 51 may be created manually by an expert.

[0059] Finally, the control unit 21 associates the created related sentence 51 with the word 41 and stores them in the storage unit 22.

[0060] In another aspect, the related sentences 51 may be generated by the related sentence data acquisition unit 2132 instead of being acquired from the related sentence database 221. The related sentence data acquisition unit 2132 generates one or more related sentences 51 that belong to the second knowledge domain 5 and have content related to the term 41, based on a term 41 that belongs to the first knowledge domain 4 specified as a search key.

[0061] In this case, the related sentence data acquisition unit 2132 may generate the related sentences 51 using a generative AI constructed using a known large-scale language model such as a Generative Pre-trained Transformer (GPT). For example, a database belonging to the second knowledge domain 5 may be searched for the phrase 41 in the first knowledge domain 4, and frequently occurring words may be extracted from the hit documents. Meaningful related sentences 51 may be created by combining the frequently occurring words, and the phrase 41 and the related sentences 51 may be associated with each other.

[0062] The user selects one related sentence 51 from the six related sentences 51 sent from the related sentence data acquisition unit 2132. The user selects at least two components, including a subject and a predicate, from among the components that make up the selected related sentence 51.

[0063] Using at least two components selected by the user, a search for literature belonging to the second knowledge domain 5, "biology," is performed. If the component is only a predicate, there may be too many hits, resulting in the inclusion of many unnecessary documents. Furthermore, if the component is three, a subject, a term, and an object, there may be too few hits. The user may change the related sentences 51 selected, increase or decrease the number of components of the selected sentence, or change the combination of components of the selected sentence, based on the search results. In this way, by appropriately selecting the number of related sentences 51 or components, an appropriate number of appropriate related sentences 51 belonging to the second knowledge domain 5 can be extracted.

[0064] As mentioned above, the elements that make up a sentence may be selected or weighted according to the purpose of the input search.

[0065] 6 is a diagram showing an example of a selection screen for related sentences 51 displayed on the display unit 13 of the user terminal 1. The user enters "1" in the related sentence selection field 51a to select the related sentence 51 in field 1. The user also checks the subject and predicate in the element selection field 51c to select the subject "plant leaves" and the predicate [resist] of the related sentence 51 in field 1. The user terminal 1 transmits "plant leaves" and [resist] to the search server 2 via the communication unit 14 as a keyword set 61 (sentence) selected by the user.

[0066] In another embodiment, the keyword set 61 may be generated without accepting user selection of related sentences 51 and / or elements. For example, the priority or weighting when selecting sentences or elements may be set depending on the purpose of the search. Also, the number of related sentences 51 associated with a phrase 41 may be specified. Limiting the number of related sentences 51 makes it easier to find specific ones, while increasing the number makes it easier to find a wider range of items. Also, the retrieved related sentences 51 may be modified by the user.

[0067] The literature search unit 2142 (search unit), which will be described later, searches for information belonging to the second knowledge domain 5, "biology," based on the keyword set 61. Here, the keyword set 61 is created from related sentences 51 belonging to the second knowledge domain 5, "biology," which is different from the first knowledge domain 4, "engineering," and which are related to the phrase 41, "water repellency," which belongs to the first knowledge domain 4, "engineering." The keyword set 61 also includes at least two of the elements of the sentences that make up the related sentences 51, and includes at least a subject and a predicate. The related sentences 51 are sentences that indicate at least one of the functions, characteristics, actions, phenomena, or states that the phrase 41 is related to in the second knowledge domain 5, "biology."

[0068] (Reference List Acquisition Unit 214) 2, the document list acquisition unit 214 includes a keyword set acquisition unit 2141 (acquisition unit), a document search unit 2142 (search unit), and a document list data generation unit 2143.

[0069] The keyword set acquisition unit 2141 acquires the keyword set 61 transmitted from the user terminal 1 via the communication control unit 216, and transmits it to the document search unit 2142. The keyword set acquisition unit 2141 may store the acquired keyword set 61 in the keyword set storage unit 222 of the storage unit 22.

[0070] The literature search unit 2142 searches for information belonging to the second knowledge domain 5, "biology," using the keyword set 61 sent from the keyword set acquisition unit 2141. Here, the literature search unit 2142 searches the field A literature database 31, which is a database that stores literature and the like belonging to biology.

[0071] The document list data generation unit 2143 generates document list data searched and extracted by the document search unit 2142, and transmits it to the output unit 215. The document list data generation unit 2143 may store the generated document list data in the document list data storage unit 223 of the storage unit 22.

[0072] The output unit 215 transmits the document list data transmitted from the document list data generation unit 2143 to the user terminal 1 via the communication control unit 216 .

[0073] FIG. 7 is a diagram showing an example of the document list data of the search results displayed on the display unit 13 of the user terminal 1. As shown in FIG. 7, the document list field 71a displays a list of documents extracted by the document search unit 2142. Along with the document list, the first knowledge domain 4, the second knowledge domain 5, the terms 41 belonging to the first knowledge domain, related sentences 51, and the keyword set 61 selected or entered by the user are also displayed. This is not limited to this, and only some of these may be displayed, or only the document list may be displayed. The output format is not limited to a display, and downloadable data may be output. A graph visualizing the search results, such as a knowledge graph, may also be displayed. Alternatively, link information for browsing documents on the web, a database generated based on the search results, highly relevant sentences extracted from the documents, etc. may also be displayed. Alternatively, numerical data of a score corresponding to the relevance of the document may be output along with the document list. For example, the higher the score, the higher the relevance with the term 41. This numerical value may also be visualized in a graph such as a knowledge graph. As yet another example, a summary of the document created by generative AI or the like may be displayed.

[0074] Although the case where one related sentence 51 is selected has been described, multiple related sentences 51 may be selected. Then, a logical sum or logical product may be created for each set obtained by the keyword sets 61 extracted from each related sentence 51.

[0075] <Processing flow> Fig. 8 is a flowchart showing an example of the flow of processing performed by the control unit 21 of the search server 2. As shown in Fig. 8, the knowledge domain acquisition unit 211 acquires information on the first knowledge domain 4 "Engineering" and the second knowledge domain 5 "Biology" transmitted from the user terminal 1 via the communication control unit 216 (S1). In addition, the control unit 21 acquires the purpose of the search transmitted from the user terminal 1 via the communication control unit 216.

[0076] Next, the term acquisition unit 212 acquires the term 41 "water repellent" belonging to the first knowledge domain 4, which is designated as a search key and which has been transmitted from the user terminal 1 via the communication control unit 216 (S2). Note that a translation of the term 41 "water repellent" into another language, or a term that is synonymous with or similar in meaning to the term 41 "water repellent", may also be designated as a search key.

[0077] The related sentence database selection unit 2131 selects the databases relating to "engineering" and "biology" acquired by the knowledge domain acquisition unit 211 from the related sentence database 221 stored in advance in the storage unit 22.

[0078] The related sentence data acquisition unit 2132 acquires from the database related sentences 51 belonging to the second knowledge domain 5 "biology" that indicate at least one of the functions, characteristics, actions, phenomena, or states related to the phrase 41 "water repellent" that belongs to the first knowledge domain 4 "engineering" acquired by the phrase acquisition unit 212, and transmits them to the user terminal 1 (S3: acquisition step). Note that the related sentence data acquisition unit 2132 may also acquire from the database related sentences 51 belonging to the second knowledge domain 5 "biology" that indicate at least one of the functions, characteristics, actions, phenomena, or states related to a translated phrase, synonymous phrase, or similar phrase of the phrase 41 "water repellent," and transmit them to the user terminal 1.

[0079] The keyword set acquisition unit 2141 acquires the keyword set 61 transmitted from the user terminal 1 via the communication control unit 216, and transmits it to the document search unit 2142 (S4). Note that the keyword set acquisition unit 2141 may acquire the keyword set 61 based on the purpose of the search acquired by the control unit 21.

[0080] The literature search unit 2142 uses the keyword set 61 sent from the keyword set acquisition unit 2141 to search for information belonging to the second knowledge domain 5, "biology" (S5: search step).

[0081] The output unit 215 outputs the document list data, which is the search result searched by the document search unit 2142, and transmits it to the user terminal 1 via the communication control unit 216 (S6: output step). Note that the output data is not limited to the document list data, and may be in other forms as described above.

[0082] In this way, by searching for information belonging to the second knowledge domain 5 based on one or more related sentences 51 belonging to the second knowledge domain 5 that are associated with words 41 belonging to the first knowledge domain 4, it is possible to link information from different knowledge domains. This improves the accuracy of the search. From another perspective, it is possible to collect information desired by the user with high accuracy. As a result, the output information group becomes a collection that better suits the user's needs.

[0083] In this way, by using the related sentence database 221 that associates the term 41 "water repellent" belonging to the first knowledge domain 4, engineering, with related sentences 51 belonging to the second knowledge domain 5, biology, which indicate the function, characteristic, action, phenomenon, or state that "water repellent" signifies, it is possible to extract information that is highly relevant in biology.

[0084] [Software implementation example] The functions of the control unit 21 (hereinafter referred to as the "device") of the search server 2 are realized by a program for causing a computer to function as the device, and by a program for causing a computer to function as each control block of the device (particularly each part included in the control unit 21).

[0085] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The control device and storage device execute the program, thereby realizing the functions described in each of the above embodiments.

[0086] The program may be non-transitory and may be recorded on one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

[0087] Furthermore, some or all of the functions of the control blocks can be realized by logic circuits. For example, an integrated circuit in which a logic circuit that functions as each of the control blocks is formed is also included in the scope of the present invention. In addition, the functions of the control blocks can also be realized by, for example, a quantum computer.

[0088] Furthermore, each process described in each of the above embodiments may be executed by AI. In this case, the AI ​​may run on the control device or on another device (for example, an edge computer or a cloud server).

[0089] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

[0090] 〔summary〕 The information processing system according to aspect 1 of the present invention includes an acquisition unit that acquires sentences that belong to a second knowledge domain different from the first knowledge domain and that are related to terms that belong to the first knowledge domain, a search unit that searches for information that belongs to the second knowledge domain based on the sentences acquired by the acquisition unit, and an output unit that outputs search results by the search unit.

[0091] According to the above aspect, it is possible to provide a search system that can search for information by linking different knowledge domains.

[0092] In the information processing system of aspect 2 of the present invention, in aspect 1 above, the sentence may be a sentence indicating at least one of the functions, characteristics, actions, phenomena, or states to which the phrase is associated in the second knowledge domain.

[0093] According to the above aspect, it is possible to search for information related to the function, characteristic, action, phenomenon, or state to which the phrase is related in the second knowledge domain.

[0094] In the information processing system of aspect 3 of the present invention, in aspect 1 or 2 above, the phrase may be a phrase indicating at least one of a function, characteristic, action, phenomenon, or state in the first knowledge domain.

[0095] According to the above aspect, by specifying a phrase indicating a function, characteristic, action, phenomenon, or state in the first knowledge domain as a search key, it is possible to search for information in the second knowledge domain that is related to the function, characteristic, action, phenomenon, or state that the phrase indicates.

[0096] An information processing system according to aspect 4 of the present invention may be such that, in any of aspects 1 to 3 above, the sentence includes at least a subject and a predicate, and the search unit searches using at least two of the elements constituting the sentence acquired by the acquisition unit.

[0097] In the information processing system according to a fifth aspect of the present invention, in any one of the first to fourth aspects, the phrase may be at least one of a single word and a compound word.

[0098] In an information processing system according to aspect 6 of the present invention, in any of aspects 1 to 5 above, the acquisition unit may acquire the sentence associated with the phrase specified as a search key from a knowledge database that stores phrases belonging to the first knowledge domain and one or more sentences belonging to the second knowledge domain that have content related to the phrases, in association with each other.

[0099] In an information processing system according to aspect 7 of the present invention, in any of aspects 1 to 5 above, the acquisition unit may generate one or more sentences belonging to the second knowledge domain, based on the phrase belonging to the first knowledge domain that is specified as a search key, and that have content related to the phrase.

[0100] An information processing system according to an eighth aspect of the present invention is the information processing system of any one of the first to seventh aspects, wherein the first knowledge domain is an engineering domain, and the second knowledge domain is a biology domain.

[0101] According to the above-mentioned aspect, it is possible to search for ideas for applying biological phenomena to engineering. That is, it is possible to search for the biological species, biological parts, biological structures (components), mechanisms, etc. that realize the function of a material to be developed.

[0102] A control method for an information processing system according to aspect 9 of the present invention is a control method for an information processing system that searches for information in a second knowledge area using information in a first knowledge area as a key, and includes an acquisition step of acquiring sentences that belong to a second knowledge area different from the first knowledge area and are related to terms that belong to the first knowledge area, a search step of searching for information that belongs to the second knowledge area based on the sentences acquired in the acquisition step, and an output step of outputting the search results obtained in the search step.

[0103] The program according to aspect 10 of the invention may be a program for causing a computer to function as the information processing system described in any one of aspects 1 to 8, and may also be a program for causing a computer to realize the functions provided by the information processing system. [Explanation of symbols]

[0104] 100 Search system (information processing system) 2132 Related sentence data acquisition unit (acquisition unit) 2141 Keyword set acquisition unit (acquisition unit) 2142 Literature Search Department (Search Department) 215 Output section 221 Related Sentence Database (Knowledge Database) 4. First Knowledge Area 41 words 5 Second Knowledge Area 51 Related sentences (sentences) 61 Keyword Set (Sentence)

Claims

1. an acquisition unit that acquires sentences that are related to a phrase that belongs to a first knowledge domain and that belong to a second knowledge domain different from the first knowledge domain; a search unit that searches for information belonging to the second knowledge domain based on the sentence acquired by the acquisition unit; an output unit that outputs the search results by the search unit, Information processing system.

2. the sentence is a sentence indicating at least one of a function, a property, an action, a phenomenon, or a state to which the phrase is related in the second knowledge domain; The information processing system according to claim 1 .

3. The phrase is a phrase that indicates at least one of a function, a property, an action, a phenomenon, or a state in the first knowledge domain. The information processing system according to claim 1 .

4. The sentence includes at least a subject and a predicate, the search unit performs a search using at least two of the elements constituting the sentence acquired by the acquisition unit; The information processing system according to claim 1 .

5. The phrase is at least one of a single word and a compound word. The information processing system according to claim 1 .

6. the acquiring unit acquires the sentence associated with the phrase designated as a search key from a knowledge database that stores a phrase belonging to the first knowledge domain and one or more sentences belonging to the second knowledge domain, the sentences having content related to the phrase, in association with each other; The information processing system according to claim 1 .

7. the acquisition unit generates one or more sentences that belong to the second knowledge domain and have content related to the phrase, based on the phrase that belongs to the first knowledge domain and is specified as a search key; The information processing system according to claim 1 .

8. the first knowledge domain is an engineering domain; the second knowledge domain is the biological domain; The information processing system according to claim 1 .

9. A control method for an information processing system that searches for information in a second knowledge domain using information in a first knowledge domain as a key, comprising: an acquisition step of acquiring sentences that belong to the second knowledge domain different from the first knowledge domain and that are related to the phrases that belong to the first knowledge domain; a search step of searching for information belonging to the second knowledge domain based on the sentence acquired in the acquisition step; an output step of outputting the search results obtained in the search step, A method for controlling an information processing system.

10. A program for causing a computer to function as the information processing system according to any one of claims 1 to 8, the program causing the computer to realize functions provided by the information processing system.

Images