Information processing system, information processing method, and program

Abstract

To provide an information processing system capable of displaying more appropriate search results.SOLUTION: An information processing system comprises a search information acquisition unit 442, an analysis unit 443, a similarity calculation unit 445, and a display unit 446. The search information acquisition unit 442 acquires a search statement of a user. The analysis unit 443 analyzes search information of the user from a plurality of perspectives by using an AI system, and creates a plurality of analyzed search statements that has been analyzed by each of the plurality of perspectives. The similarity calculation unit 445 calculates, for each of a plurality of pieces of retained information stored in a database, a similarity of retained information with respect to a plurality of pieces of analyzed search information. The display unit 446 displays the plurality of pieces of retained information on a terminal device in order of highest similarity as search results for the search information of the user.SELECTED DRAWING: Figure 5

Description

Technical Field

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

Background Art

[0002] Conventionally, there is a search method described in Patent Document 1 below. In this search method, a plurality of consecutive words are partially cut out from an input character string composed of a plurality of words and blank characters, and one to a plurality of alternative characters are applied to two or more consecutive characters to obtain a search key character string. By using this search key character string for searching, it is possible to extract those that are more likely to match than a search target character string with a plurality of spelling mistakes.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the search method described in Patent Document 1, there is room for improvement regarding the output of search results. The present invention has been made in view of such circumstances, and an object thereof is to provide an information processing system, an information processing method, and a program capable of displaying more appropriate search results.

Means for Solving the Problems

[0005] The information processing system for solving the above problems includes a search information acquisition unit, an analysis unit, a similarity calculation unit, and a display unit. The search information acquisition unit acquires the user's search information from the terminal device. The analysis unit analyzes the user's search information from multiple perspectives by an artificial intelligence system and creates multiple pieces of analyzed search information analyzed by each of the multiple perspectives. The similarity calculation unit calculates the similarity of the possession information with respect to the multiple pieces of analyzed search information for each of the multiple pieces of possession information stored in the database. The display unit causes the terminal device to display, in descending order of similarity, the multiple pieces of possession information or information related to each of the multiple pieces of possession information as search results for the user's search information.

[0006] In the information processing method for solving the above problems, a computer acquires the user's search information from the terminal device, analyzes the user's search information from multiple perspectives by an artificial intelligence system, creates multiple pieces of analyzed search information analyzed by each of the multiple perspectives, calculates the similarity of the possession information with respect to the multiple pieces of analyzed search information for each of the multiple pieces of possession information stored in the database, and causes the terminal device to display, in descending order of similarity, the multiple pieces of possession information or information related to each of the multiple pieces of possession information as search results for the user's search information.

[0007] In the program for solving the above problems, the computer is caused to acquire the user's search information from the terminal device, analyze the user's search information from multiple perspectives by an artificial intelligence system, create multiple pieces of analyzed search information analyzed by each of the multiple perspectives, calculate the similarity of the possession information with respect to the multiple pieces of analyzed search information for each of the multiple pieces of possession information stored in the database, and cause the terminal device to display, in descending order of similarity, the multiple pieces of possession information or information related to each of the multiple pieces of possession information as search results for the user's search information.

[0008] According to this configuration, after the search information input by the user is analyzed by the artificial intelligence system, the similarity of the held information is calculated for the plurality of analyzed search information that is the analysis result, so that it is possible to calculate a higher-precision similarity. And according to the above configuration, in the terminal device, as search results, a plurality of held information in descending order of similarity, or information related to each of the plurality of held information is displayed, so that it is possible to display more appropriate search results.

Effect of the Invention

[0009] According to the information processing system, information processing method, and program of the present invention, more appropriate search results can be displayed.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Mode for Carrying Out the Invention

[0011] Hereinafter, an embodiment of an information processing system, an information processing method, and a program will be described with reference to the drawings. To facilitate understanding of the description, the same reference numerals are given to the same components in each drawing as much as possible, and duplicate descriptions are omitted. <Embodiment> First, an overview of the information processing system according to the embodiment will be described.

[0012] (Overview of Information Processing System) The information processing system 10 shown in FIG. 1 is a system that displays search results corresponding to search information input by a user. As shown in FIG. 1, the information processing system 10 includes a terminal device 20, a first server device 30, and a second server device 40.

[0013] The terminal device 20 is a portable smart device such as a smartphone or a tablet, or a personal computer (such as a desktop, laptop, tablet, etc.) installed at a predetermined location. The server devices 30 and 40 are, for example, cloud computers, server computers, personal computers, etc. The terminal device 20, the first server device 30, and the second server device 40 are communicably connected to each other via the communication network 11. The communication network 11 may be, for example, a LAN, a dedicated line, a telephone line, a mobile communication network, WiFi (Wireless Fidelity), any other communication line, or a combination thereof, regardless of whether it is wired or wireless.

[0014] The terminal device 20 is possessed by a user who uses the information processing system 10. By operating the terminal device 20, the user can view the web pages provided by the first server device 30 and the second server device 40, respectively. When the first server device 30 and the second server device 40 are requested for web page information from the terminal device 20 via a browser or the like, they transmit the requested web page information to the terminal device 20. As a result, on the terminal device 20, the web pages transmitted from the first server device 30 and the second server device 40 are displayed on the browser. The first server device 30 provides a predetermined web page to the user. In this embodiment, a case where the web page provided to the user by the first server device 30 is a web page that displays cooking recipes in text information and image information will be described as an example. The second server device 40 provides a web page to the user that enables searching for a plurality of cooking recipes held by the first server device 30.

[0015] (Configuration of the terminal device) Next, the configuration of the terminal device 20 will be described. As shown in FIG. 2, the terminal device 20 includes a storage unit 21, a communication unit 22, an input unit 23, an output unit 24, and a control unit 25.

[0016] The storage unit 21 stores various types of information possessed by the terminal device 20. For example, the storage unit 21 stores various programs for operating the terminal device 20. The communication unit 22 performs various communications with the first server device 30, the second server device 40, etc. via the communication network 11.

[0017] The input unit 23 has a function of receiving the user's input operations, such as a mouse, a keyboard, operation buttons, etc. The output unit 24 has a function of displaying various screens, such as a liquid crystal display, etc. Hereinafter, a case where the input unit 23 and the output unit 24 are configured by a touch pad in which their functions are integrated will be taken as an example for explanation.

[0018] The control unit 25 comprehensively controls the terminal device 20. The control unit 25 has, for example, a display unit 250 as a functional configuration realized by executing the program stored in the storage unit 21. The display unit 250 displays a predetermined web page on the browser based on, for example, the web page information transmitted from the first server device 30 and the second server device 40.

[0019] (Configuration of the First Server Device) Next, the configuration of the first server device 30 will be described. As shown in FIG. 3, the first server device 30 includes a storage unit 31, a communication unit 32, and a control unit 33.

[0020] The storage unit 31 stores various types of information possessed by the first server device 30. For example, the storage unit 31 stores a web information database 310. The web information database 310 stores web page information corresponding to each of a plurality of cooking recipes. The web page information includes URL information, as well as character information and image information of the cooking recipes, etc. The character information of the cooking recipe includes, for example, character information such as the name of the dish, the outline of the dish, the name and gram amount of the ingredients of the dish, the cooking method, etc. The image information of the cooking recipe includes an image of the dish, etc.

[0021] The communication unit 32 conducts various communications with the terminal device 20 and the second server device 40 via the communication network 11. The control unit 33 comprehensively controls the first server device 30. The control unit 33 includes, for example, a transmission unit 330 as a functional configuration realized by executing a program stored in the storage unit 31.

[0022] The transmission unit 330 transmits various information to the terminal device 20 and the second server device 40 via the communication unit 32. For example, the transmission unit 330 transmits Web page information of a cooking recipe to the terminal device 20 in response to a request from the terminal device 20. For example, assume that a user performs an operation on the terminal device 20 to display a Web page of a predetermined cooking recipe among a plurality of cooking recipes. At this time, the terminal device 20 requests the Web page information of the predetermined cooking recipe from the first server device 30. In this case, the transmission unit 330 reads the Web page information corresponding to the predetermined cooking recipe requested from the terminal device 20 from the Web information database 310 and transmits the read Web page information to the terminal device 20. As a result, a screen 80 of a Web page of a cooking recipe as shown in FIG. 4, for example, is displayed on the terminal device 20.

[0023] (Configuration of the Second Server Device) Next, the configuration of the second server device 40 will be described. As shown in FIG. 5, the second server device 40 includes a storage unit 41, a communication unit 42, and a control unit 43.

[0024] The storage unit 41 stores various information possessed by the second server device 40. For example, the storage unit 41 stores an acquisition information database 410 and a vector information database 411. As shown in FIG. 6(A), the acquired information database 410 stores text information, image information, and URL information included in each Web page of a plurality of cooking recipes held by the first server device 30. The text information is obtained by converting the character information included in the Web page of the cooking recipe into text. Therefore, the text information includes, for example, text-converted character information such as the name of the dish, the content of the dish, the name and gram amount of the ingredients of the dish, and the cooking method. In the acquired information database 410, the text information, image information, and URL information of each of the plurality of cooking recipes are stored in association with individual identification numbers. In the present embodiment, the text information, image information, URL information, identification numbers, etc. of the cooking recipes stored in the acquired information database 410 correspond to the held information.

[0025] As shown in FIG. 6(B), the vector information database 411 stores vector information obtained by vectorizing the text information of each of the plurality of cooking recipes stored in the acquired information database 410. The vector information is, for example, a numerical value obtained by quantifying the text information of the cooking recipe by means of Bag of Words, distributed representation, or the like. In the vector information database 411, the vector information of each of the plurality of cooking recipes is stored in association with individual identification numbers.

[0026] The communication unit 42 shown in FIG. 5 performs various communications with the terminal device 20 and the first server device 30 via the communication network 11. The control unit 43 comprehensively controls the second server device 40. The control unit 43 includes, for example, a search screen display unit 430 and a search processing unit 431 as a functional configuration realized by executing a program stored in the storage unit 41.

[0027] (Search screen display unit) The search screen display unit 430 transmits search Web page information to the terminal device 20 in response to a request from the terminal device 20. As a result, a search screen 50 as shown in FIG. 7, for example, is displayed on the terminal device 20.

[0028] As shown in FIG. 7, a search sentence input area 51 is provided in the search screen 50. The search sentence input area 51 is an area where the user inputs a search sentence. The search sentence that can be input into the search sentence input area 51 is not limited to a sentence that can specify the search target, and may be an ambiguous sentence that is difficult to specify the search target. As an example of such a search sentence, there is a sentence such as "I want to eat a refreshing lunch". This sentence only states the user's desire and it is difficult to specify the cooking recipe of the search target. In the information processing system 10 of the present embodiment, even when such a sentence is input as a search sentence, it is possible to provide the user with an appropriate cooking recipe for the search sentence.

[0029] (Search processing unit) The search processing unit 431 shown in FIG. 5 executes a process of searching for a cooking recipe corresponding to the user's search sentence from among a plurality of cooking recipes. The search processing unit 431 includes a held information acquisition unit 440, a first vector processing unit 441, a search information acquisition unit 442, an analysis unit 443, a second vector processing unit 444, a similarity calculation unit 445, a display unit 446, and a transmission unit 447. Next, these configurations will be specifically described with reference to FIG. 8.

[0030] (Held information acquisition unit) The owned information acquisition unit 440 acquires the Web page information of each of the plurality of cooking recipes held by the first server device 30. At this time, the owned information acquisition unit 440 acquires the text information of each Web page of the plurality of cooking recipes by performing a process of converting the Web page information of each of the plurality of cooking recipes into text information. In addition, the owned information acquisition unit 440 also acquires the image information and URL information included in the Web page information of each of the plurality of cooking recipes. When the owned information acquisition unit 440 acquires the text information, image information, and URL information of each of the plurality of cooking recipes, the acquired text information, image information, and URL information of the cooking recipes are associated with individual identification numbers and stored in the storage unit 41. Thereby, an acquired information database 410 as shown in FIG. 6(A) is constructed.

[0031] Note that the owned information acquisition unit 440 may acquire the text information, image information, and URL information of each of the plurality of cooking recipes from the first server device 30 by using an API or the like. (First vector processing unit) The first vector processing unit 441 shown in FIG. 8 executes a process of converting the text information of each of the plurality of cooking recipes stored in the acquired information database 410 into vector information. For example, the first vector processing unit 441 uses a method such as Bag of Words or distributed representation for the text information of each of the plurality of cooking recipes stored in the acquired information database 410 to convert the text information of each of the plurality of cooking recipes from string information into vector information, which is numerical information. After the first vector processing unit 441 converts the text information of each of the plurality of cooking recipes into vector information, the converted vector information is associated with an individual identification number and stored in the storage unit 41. Thereby, a vector information database 411 as shown in FIG. 6(B) is constructed.

[0032] (Search information acquisition unit) When the user inputs a search query on the search screen 50 as shown in FIG. 7, the search information acquisition unit 442 shown in FIG. 8 acquires the information of the input search query from the terminal device 20.

[0033] (Search Information Analysis Unit) The analysis unit 443 shown in FIG. 8 acquires the information of the user's search query from the search information acquisition unit 442 and analyzes the acquired search query information using various artificial intelligence systems 443a. Hereinafter, the artificial intelligence system 443a is referred to as "AI (Artificial Intelligence) system 443a". Note that the AI system 443a may be installed in the second server device 40 or may be provided by a device different from the second server device 40. The AI system 443a is an interactive AI system or the like that can communicate in a chat format. The interactive AI system is, for example, ChatGPT.

[0034] For example, when the analysis unit 443 acquires a search query such as "I want to eat a refreshing lunch" by the search information acquisition unit 442, the analysis unit 443 converts the search query into a data format analyzable by the AI system 443a. For example, the analysis unit 443 creates analysis data Da10 as shown in FIG. 9 from this search query. In the analysis data Da10, the character string "Search query:" is added to the head of the text information Dt10 of the search query, and the character string C11 of "Request: ···" is added after the search query. The character string of "Request: ···" corresponds to a prompt sentence for the AI system 443a. This prompt sentence requests the AI system 443a to predict the emotions of the user's deep psychology from the text information Dt10 of the search query.

[0035] When such analysis data Da10 is input into the AI system 443a, the AI system 443a analyzes the content of the text information Dt10 of the search sentence based on the prompt sentence "Request: ···" included in the analysis data Da10, and outputs the analysis result as text information Da11 as shown in, for example, FIG. 9. As shown in FIG. 9, the text information Da11 of the search sentence after this analysis includes the feelings of the user's deep psychology predicted by the AI system 443a from the content of the text information Dt10 of the search sentence based on the prompt sentence "Request: ···". The AI system 443a analyzes the information of the search sentence from multiple perspectives. In FIG. 9, the case where the AI system 443a analyzes the search sentence from multiple perspectives of "type of cuisine", "main ingredients", "cooking method", "flavor", and "purpose" is illustrated. In the text information Da11 of the search sentence after analysis shown in FIG. 9, the sentences described following the character strings of "type of cuisine", "main ingredients", "cooking method", "flavor", and "purpose" are the text information Dt20 to Dt24 of the search sentence after analysis obtained by the AI system 443a analyzing the text information Dt10 of the search sentence. That is, the text information Dt20 to Dt24 of the search sentence after analysis shows the results of the AI system 443a analyzing the search sentence from multiple perspectives of "type of cuisine", "main ingredients", "cooking method", "flavor", and "purpose" respectively.

[0036] Note that the perspectives used by the AI system 443a for analyzing the search sentence are not limited to the perspectives shown in FIG. 9 and can be changed as appropriate. The perspectives used by the AI system 443a for analysis may be predetermined, or may be appropriately selected by the AI system 443a according to the search sentence. (Second Vector Processing Unit) The second vector processing unit 444 shown in FIG. 8 acquires the text information Dt20 to Dt24 of the analyzed search sentence from the analysis unit 443, and performs a process of converting the acquired text information Dt20 to Dt24 of the analyzed search sentence into vector information. For example, the second vector processing unit 444 uses a method such as Bag of Words or distributed representation for the text information Dt20 to Dt24 of the analyzed search sentence transmitted from the analysis unit 443, and converts the text information Dt20 to Dt24 of the analyzed search sentence from string information into vector information, which is numerical information, to create the vector information Dt30 to Dt34 of the analyzed search sentence.

[0037] (Similarity calculation unit) The similarity calculation unit 445 acquires the vector information Dt30 to Dt34 of the analyzed search sentence from the second vector processing unit 444, and calculates the similarity between the vector information of the cooking recipe and the vector information Dt30 to Dt34 of the analyzed search sentence respectively.

[0038] For example, the similarity calculation unit 445 reads the vector information of a predetermined cooking recipe among a plurality of cooking recipes from the vector information database 411, and calculates the similarity between the read vector information of the predetermined cooking recipe and the vector information Dt30 to Dt34 of the analyzed search sentence. The similarity calculated by the similarity calculation unit 445 is, for example, the cosine similarity. The cosine similarity is represented by a value in the range from "-1" to "1". The closer the value of the cosine similarity is to "1", the higher the similarity indicates. Also, the closer the value of the cosine similarity is to "-1", the lower the similarity indicates.

[0039] The similarity calculation unit 445 calculates the similarity between the vector information of each cooking recipe stored in the vector information database 411 and the vector information Dt30 to Dt34 of the analyzed search sentence respectively. (Display unit) The display unit 446 generates information to be displayed on the terminal device 20 as search results based on the similarity of each of the plurality of cooking recipes calculated by the similarity calculation unit 445. For example, the display unit 446 acquires the text information and image information of the plurality of cooking recipes, and generates the heading information of each of the plurality of cooking recipes based on the text information and image information. The heading information is information that extracts only a part of the text information and image information of the cooking recipe, or information that summarizes the text information and image information of the cooking recipe. In the present embodiment, the heading information is an example of information related to the held information. The display unit 446 rearranges the heading information of each of the plurality of cooking recipes in descending order of similarity, and then transmits the rearranged heading information of the plurality of cooking recipes to the terminal device 20. As a result, a search result screen 60 as shown in FIG. 10 is displayed on the terminal device 20.

[0040] As shown in FIG. 10, a heading information display area 61 is provided on the search result screen 60. In the heading information display area 61, the heading information 610, 611 of each of the plurality of cooking recipes is displayed vertically in descending order of similarity. Although only the heading information 610, 611 of two cooking recipes is displayed on the search result screen 60 shown in FIG. 10, the user can further display the heading information of cooking recipes with lower similarity by, for example, swiping the heading information display area 61 upward.

[0041] (Transmission unit) When the user selects any one of the plurality of heading information on the search result screen 60 shown in FIG. 10, the transmission unit 447 transmits the URL information of the cooking recipe corresponding to the selected heading information to the terminal device 20.

[0042] Specifically, each of the heading information 610 and 611 is associated with the actual URL information of the corresponding cooking recipe, that is, the URL information of the cooking recipe stored in the web information database 310 of the first server device 30. For example, if the user performs an operation of selecting the heading information 610 on the terminal device 20, the transmission unit 447 transmits the URL information of the cooking recipe on the first server device 30 corresponding to the heading information 610 to the terminal device 20. Thereby, the terminal device 20 accesses the web page of the cooking recipe on the first server device 30 corresponding to the heading information 610. Thereby, the terminal device 20 displays the web page information of the actual cooking recipe as shown in, for example, FIG. 4.

[0043] (Operation example of the information processing system) Next, with reference to FIGS. 11 and 12, an operation example of the information processing system 10 of the present embodiment will be described. As shown in FIG. 11, in the information processing system 10 of the present embodiment, first, the possession information acquisition unit 440 of the second server device 40 acquires the text information, image information, URL information, etc. of each web page of a plurality of cooking recipes from the first server device 30 (step S30), and constructs an acquisition information database 410 (step S31). Subsequently, the first vector processing unit 441 of the second server device 40 constructs a vector information database 411 by converting the text information of each web page of a plurality of cooking recipes into vector information (step S32). Thus, the preparation for searching for a plurality of cooking recipes in the second server device 40 is completed.

[0044] Thereafter, as shown in FIG. 12, when the user operates the terminal device 20 to access the top page of the website provided by the first server device 30 (step S10), the first server device 30 transmits the information of the top page of the website to the terminal device 20 (step S20). As a result, the terminal device 20 displays, for example, the top screen 70 of the website as shown in FIG. 12 on the browser (step S11). As shown in FIG. 13, the top screen 70 is provided with a search area 71 and a concierge button 72. The search area 71 is provided with various input items used when searching for cooking recipes using the normal search system provided by the first server device 30. The concierge button 72 is a button to be operated when searching for cooking recipes using the search system provided by the second server device 40. The URL of the search web page provided by the second server device 40 is associated with the concierge button 72.

[0045] If the user operates the concierge button 72 on the top screen 70 (step S12), the terminal device 20 accesses the second server device 40 based on the URL associated with the concierge button 72 (step S13). As a result, the search screen display unit 430 of the second server device 40 transmits search Web page information to the terminal device 20 (step S33). Thereby, the terminal device 20 displays a search screen 50 as shown in FIG. 7, for example (step S14). When the user inputs a search query on this search screen 50, the information of the search query is acquired by the search information acquisition unit 442 of the second server device 40 (step S34). Subsequently, the analysis unit 443, the second vector processing unit 444, and the similarity calculation unit 445 of the second server device 40 execute a process of searching for a cooking recipe corresponding to the acquired user's search query from among a plurality of cooking recipes (step S35). Then, the display unit 446 of the second server device 40 transmits the search result information to the terminal device 20 (step S36). Thereby, the terminal device 20 displays a search result screen 60 as shown in FIG. 10, for example (step S16). If the user performs an operation of selecting, for example, the heading information 610 on this search result screen 60 (step S17), the transmission unit 447 of the terminal device 20 accesses the URL of the Web page of the first server device 30 corresponding to the heading information 610 (step S18). As a result, the first server device 30 transmits the information of the Web page of the cooking recipe corresponding to the heading information 610 to the terminal device 20 (step S21). Therefore, the terminal device 20 displays the Web page of the cooking recipe corresponding to the heading information 610 (step S19).

[0046] (Hardware configuration of the information processing system) Next, the hardware configurations of the terminal device 20, the first server device 30, and the second server device 40 will be described respectively. FIG. 14 shows the hardware configuration of a computer 100 capable of realizing the terminal device 20, the first server device 30, and the second server device 40. The computer 100 has, for example, a processor 101, a storage device 102, a communication IF (Interface) 103, an input device 104, and an output device 105. The processor 101 is a CPU (Central Processing Unit), a GPU (Graphical processing unit), or the like. The storage device 102 is composed of, for example, at least one of a memory, an HDD (Hard Disk Drive), and an SSD (Solid State Drive). The communication IF 103 performs wired communication or wireless communication. The input device 104 is a device that receives an input operation and is composed of, for example, at least one of a keyboard, a touch panel, a mouse, and a microphone. The output device 105 is a device that outputs information and is composed of, for example, at least one of a display, a touch panel, and a speaker.

[0047] (Operation and Effect of Information Processing System) As described above, the information processing system 10 of the present embodiment includes a search information acquisition unit 442, an analysis unit 443, a similarity calculation unit 445, and a display unit 446. The search information acquisition unit 442 acquires a user's search query (search information) from the terminal device 20. The analysis unit 443 analyzes the user's search information from a plurality of viewpoints by the AI system 443a and creates text information (analyzed search information) Dt20 to Dt24 of the analyzed search query analyzed by each of the plurality of viewpoints. The similarity calculation unit 445 calculates the similarity of each recipe information (held information) of a plurality of cooking recipes stored in the acquisition information database 410 with respect to the text information Dt20 to Dt24 of the plurality of analyzed search queries. The display unit 446 causes the terminal device 20 to display, in descending order of similarity, the heading information of each of the plurality of cooking recipes (information corresponding to each of the plurality of held information) as a search result for the user's search query.

[0048] According to this configuration, after the search query input by the user is analyzed by the AI system 443a, the similarity of cooking recipes to the text information Dt20 to Dt24 of a plurality of post-analysis search queries, which is the analysis result, is calculated. Therefore, it is possible to calculate a higher-precision similarity. And according to the above configuration, on the terminal device 20, a plurality of cooking recipes are displayed as search results in descending order of similarity. Therefore, it is possible to display more appropriate search results.

[0049] The information of the cooking recipes stored in the acquisition information database 410 includes the text information (character information) in the Web page. The similarity calculation unit 445 calculates the similarity of the text information of each of the plurality of cooking recipes stored in the acquisition information database 410 to the text information Dt20 to Dt24 of the plurality of post-analysis search queries respectively.

[0050] According to this configuration, it is possible to calculate a higher-precision similarity. The information processing system 10 further includes a first vector processing unit 441 and a second vector processing unit 444. The first vector processing unit 441 converts the text information of the Web pages of the plurality of cooking recipes stored in the acquisition information database 410 into vector information. The second vector processing unit 444 converts the text information Dt20 to Dt24 of the plurality of post-analysis search queries into vector information. When the similarity calculation unit 445 calculates the similarity of each of the plurality of cooking recipes to the text information Dt20 to Dt24 of the plurality of post-analysis search queries, it uses the vector information of the plurality of post-analysis search queries and the text information of each of the Web pages of the plurality of cooking recipes.

[0051] According to this configuration, it is possible to calculate a higher-precision similarity. The information of the cooking recipes stored in the acquisition information database 410 further includes the URL information of the Web page in association with the text information (character information) in the Web page. The information processing system 10 further includes a transmission unit 447. When any one of the heading information of a plurality of cooking recipes displayed on the terminal device 20 is selected by the user, the transmission unit 447 transmits the URL information of the Web page corresponding to the selected cooking recipe to the terminal device 20.

[0052] According to this configuration, the user can view the actual cooking recipe Web page held by the first server device 30 only by selecting any one of the heading information of a plurality of cooking recipes, so that the convenience can be improved. When the user performs an operation requesting a search on the terminal device 20, the display unit 446 further causes the terminal device 20 to display a search screen 50. The search information acquisition unit 442 acquires the information input by the user on the search screen 50 as the user's search information.

[0053] According to this configuration, it becomes possible to easily acquire the information that the user wants to search for. (First Modification Example) Next, a first modification example of the information processing system 10 of the embodiment will be described.

[0054] As shown in FIG. 15, an image attachment button 52 is further provided on the search screen 50 of this modification example. The user can also search for the cooking recipe to be searched by image information by operating the image attachment button 52. Hereinafter, the image attached by operating the image attachment button 52 is also referred to as a "search image". In this modification, the user can select any one of a search using only the search sentence input in the search sentence input area 51, a search using only the search image, and a search using both the search sentence and the search image.

[0055] For example, when the user inputs a search query in the search query input area 51 and attaches a search image by operating the image attachment button 52, the search query and the search image are acquired by the search information acquisition unit 442 of the second server device 40. In this case, the AI system 443a of the analysis unit 443 creates text information of the analyzed search query corresponding to each of the plurality of viewpoints by analyzing the search query and the search image from a plurality of viewpoints.

[0056] According to this configuration, the user can use not only text information but also image information as search information, so that the convenience can be improved. (Second Modified Example) Next, a second modified example of the information processing system 10 of the embodiment will be described.

[0057] FIG. 16 schematically shows the information included in the acquired information database 410 of this modified example. As shown in FIG. 16, in the acquired information database 410, flag information F1, F2,... is further stored in association with the identification number, the text information of the Web page of the cooking recipe, the image information, and the URL information of the Web page. Each flag information F1, F2,... indicates the attributes of the cooking recipe. For example, the flag information F1 is information indicating whether the recipe is related to the "autumn season". Also. The flag information F2 is information indicating whether the recipe is related to the "spring season". The flag information F1, F2,... is set to "1" when it is in the on state and set to "0" when it is in the off state. For example, when the flag information F1 corresponding to a predetermined recipe is set to "1", that is, when the flag information F1 is in the on state, it indicates that the predetermined recipe is related to the autumn season.

[0058] Note that each flag information F1, F2,... is set by the inventor of the cooking recipe and is stored in the Web information database 310 of the first server device 30. The possession information acquisition unit 440 of the second server device 40 acquires each flag information F1, F2,... of a plurality of cooking recipes from the first server device 30 and stores them in the acquired information database 410.

[0059] The display unit 446 of the second server device 40 determines whether specific search information corresponding to each flag information F1, F2, ··· is included in the user's search information. For example, regarding the flag information F1, the word "autumn" is set in the specific search information. Also, regarding the flag information F2, the word "spring" is set in the specific search information. Therefore, the display unit 446 determines whether the word "autumn" corresponding to the flag information F1 or the word "spring" corresponding to the flag information F2 is included in the user's search sentence. For example, when the word "autumn" is included in the user's search sentence, the display unit 446 preferentially displays the cooking recipes with the flag information F1 set to the on state on the search result screen 60. For example, the display unit 446 arranges and displays the heading information of each of the multiple cooking recipes with the flag information F1 set to the on state in descending order of similarity, and below that, arranges and displays the heading information of each of the multiple cooking recipes with the flag information F1 set to the off state in descending order of similarity.

[0060] In this way, the display unit 446 of this modification example changes the order of arrangement of the heading information of each of the multiple cooking recipes to be displayed on the terminal device 20 based on the flag information F1, F2, ···. According to this configuration, the intention of the inventor of the cooking recipe can be further reflected in the search results. <Other Embodiments> The present disclosure is not limited to the above specific examples.

[0061] The information held by the first server device 30 and the information held by the information acquisition unit 440 of the second server device 40 that is acquired from the first server device 30 are not limited to the Web page information of cooking recipes, and any information can be used. For example, those pieces of held information may be information of a Web page on which some points can be acquired. Further, those pieces of information are not limited to Web pages, and may be, for example, information of various books, various document information held by a company, and the like. In that case, on the search result screen 60, instead of the heading information summarizing the information held by the first server device 30, the information held by the first server device 30 may be displayed as it is. That is, the display unit 446 may display a plurality of pieces of held information as they are in descending order of similarity to the user's search query on the terminal device 20 as search results for the user's search query.

[0062] The information held by the first server device 30 and the information held by the information acquisition unit 440 of the second server device 40 that is acquired from the first server device 30 may be information of a Web page of a travel site. According to this configuration, for example, when a user inputs a search query including the time of travel, the destination, the purpose of travel, etc. on the search screen 50, it is possible to realize a system that proposes recommended travel destinations.

[0063] The information held by the first server device 30 and the information held by the information acquisition unit 440 of the second server device 40 that is acquired from the first server device 30 may be information of e-books. According to this configuration, for example, when a user inputs a search query including the user's past reading history, the characteristics of the user's favorite works, etc. on the search screen 50, it is possible to realize a system that proposes recommended e-books.

[0064] The configuration of the information processing system 10 of this embodiment can also be used for, for example, searches within an EC site, searches within a news site, searches within a video site, searches within an SNS (Social Networking Service) site, searches within a specific information site, searches within a portal site, and searches within a corporate site. A specific information site is a site created for a specific purpose, such as a job hunting / employment site, a car information site, a restaurant site, and a travel site.

[0065] Those in which those skilled in the art have appropriately made design changes to the above specific examples are also included in the scope of the present disclosure as long as they have the features of the present disclosure. Each element, and its arrangement, conditions, shape, etc. provided in each of the above specific examples are not limited to those illustrated and can be appropriately changed. Each element provided in each of the above specific examples can be appropriately combined as long as there is no technical contradiction.

Description of Reference Numerals

[0066] 10: Information processing system, 100: Computer, 441: First vector processing unit, 442: Search information acquisition unit, 443: Analysis unit, 444: Second vector processing unit, 445: Similarity calculation unit, 446: Display unit, 447: Transmission unit.

Claims

1. a search information acquisition unit that acquires user search information from a terminal device; an analysis unit that analyzes the user's search information from a plurality of perspectives using an artificial intelligence system and creates a plurality of pieces of analyzed search information analyzed from each of the plurality of perspectives; a similarity calculation unit that calculates a similarity of each of a plurality of pieces of retained information stored in a database with respect to the plurality of pieces of post-analysis search information; a display unit that displays, on the terminal device, the plurality of pieces of held information or information related to each of the plurality of pieces of held information in order of the degree of similarity as a search result for the user's search information. Information processing system.

2. The retained information includes text information in a predetermined web page, The similarity calculation unit calculates a similarity of the character information with respect to each of the plurality of pieces of retained information stored in the database, with respect to the plurality of pieces of post-analysis search information. The information processing system according to claim 1 .

3. a first vector processing unit that converts character information in each of the plurality of web pages stored in the database into vector information; A second vector processing unit that converts each of the plurality of pieces of post-analysis search information into vector information, The similarity calculation unit uses the plurality of pieces of post-analysis search information converted into vector information and character information of each of the plurality of web pages converted into vector information when calculating the similarity of the retained information to each of the plurality of pieces of post-analysis search information. The information processing system according to claim 2 .

4. The retained information further includes URL information of the specified web page in association with character information in the specified web page, a transmission unit that transmits, when the user selects any one of the plurality of pieces of held information displayed on the terminal device, or when the user selects any one of pieces of information related to each of the plurality of pieces of held information displayed on the terminal device, to the terminal device URL information of the predetermined Web page corresponding to the selected held information. The information processing system according to claim 2 .

5. The retained information further includes flag information that can be set to indicate whether or not specific search information is to be preferentially displayed in the search results, The display unit changes an order of the plurality of pieces of retained information or information related to each of the plurality of pieces of retained information to be displayed on the terminal device based on the flag information. The information processing system according to claim 1 .

6. the display unit further displays a search screen on the terminal device when the user performs an operation on the terminal device to request a search, The search information acquisition unit acquires information input by the user on the search screen as the user's search information. The information processing system according to claim 1 .

7. The similarity calculation unit uses a cosine similarity as the similarity. The information processing system according to claim 1 .

8. The search information acquisition unit acquires at least one of text information and image information from the terminal device as the user's search information. The information processing system according to claim 1 .

9. The computer Acquire user search information from the terminal device; Analyzing the user's search information from a plurality of perspectives using an artificial intelligence system to generate a plurality of pieces of analyzed search information analyzed from each of the plurality of perspectives; Calculating a similarity of each of a plurality of pieces of retained information stored in a database with respect to the plurality of pieces of post-analysis retrieved information; The plurality of pieces of held information or pieces of information related to each of the plurality of pieces of held information are displayed on the terminal device in order of the degree of similarity as a search result for the user's search information. Information processing methods.

10. On the computer, Acquire user search information from the terminal device; Analyzing the user's search information from a plurality of perspectives using an artificial intelligence system to generate a plurality of pieces of analyzed search information analyzed from each of the plurality of perspectives; Calculating a similarity of each of a plurality of pieces of retained information stored in a database with respect to the plurality of pieces of post-analysis retrieved information; The terminal device displays the plurality of pieces of held information or pieces of information related to each of the plurality of pieces of held information in order of the degree of similarity as a search result for the user's search information. program.

Images