Information processing device, information processing method, and information processing program

Abstract

To correct a search query to an appropriate search query with less excessive corrections.SOLUTION: An information processing apparatus according to the present invention includes: an estimation unit which, when accepting a search query including a plurality of keywords preliminarily set as non-correction objects, estimates a degree of association between the keywords; and a correction unit which does not correct the search query when the degrees of association estimated by the estimation unit exceeds a threshold, and corrects the search query when the degrees of association is lower than the threshold.SELECTED DRAWING: Figure 1

Description

Technical Field

[0004] ,

[0006] , , , , , ,

[0005] , , , , ,

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

Background Art

[0002] Conventionally, techniques for correcting (converting) keywords input by a user have been provided. For example, when a search query input by a user includes a spelling mistake or the like, a technique for correcting the spelling mistake and then providing a search result to the user has been provided (for example, Patent Document 1 and the like).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the prior art, there are cases where the correction is not made to an appropriate search query. For example, when correcting a search query using a model that has performed machine learning, the search query may be excessively corrected. Further, in order to suppress the excessive correction of the search query by the above model, a method of registering in advance keywords that are not targets for correction can be considered, but there is a risk that a query that should originally be corrected cannot be corrected.

[0005] The present invention has been made in view of the above, and an object thereof is to provide an information processing apparatus, an information processing method, and an information processing program that can correct to an appropriate search query while suppressing excessive correction. s

Means for Solving the Problems

[0006] To solve the above-mentioned problems and achieve the objective, the information processing device according to the present invention includes an estimation unit that estimates the degree of relevance between keywords when it receives a search query containing a plurality of keywords that have been set in advance as to be uncorrected, and a correction unit that does not correct the search query if the degree of relevance estimated by the estimation unit exceeds a threshold, and corrects the search query if the degree of relevance falls below the threshold. [Effects of the Invention]

[0007] According to the present invention, it is possible to correct to an appropriate search query while suppressing excessive correction. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 shows an example of information processing according to the embodiment. [Figure 2] Figure 2 is a block diagram showing an example configuration of an information providing device according to an embodiment. [Figure 3] Figure 3 shows an example of a search log storage unit according to the embodiment. [Figure 4] Figure 4 is a schematic diagram of the estimation process according to the embodiment. [Figure 5] Figure 5 is a flowchart showing an example of the correction process according to this embodiment. [Figure 6] Figure 6 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device according to the embodiment. [Modes for carrying out the invention]

[0009] The following describes in detail, with reference to the drawings, the embodiments for implementing the information processing device, information processing method, and information processing program relating to this application (hereinafter referred to as "embodiments"). However, these embodiments do not limit the information processing device, information processing method, and information processing program relating to this application.

[0010] [Embodiment] [1. Information Processing] First, an example of information processing according to the embodiment will be described using Figure 1. Figure 1 is a diagram showing an example of information processing according to the embodiment. The information processing according to the embodiment is realized by the information processing device 1 shown in Figure 1.

[0011] Information processing device 1 is a server device that provides various search services. For example, information processing device 1 uses search queries entered from user terminal 50 to search for content to be searched and provides the search results to user terminal 50.

[0012] In this process, the information processing device 1 corrects any typographical errors in keywords included in the search query received from the user terminal 50 using a correction model. The correction model is, for example, a machine learning model that uses the keyword to be corrected and its correct equivalent keyword as ground truth data.

[0013] Here, the correction model may perform overcorrection, correcting keywords that do not actually need to be corrected. As a countermeasure against overcorrection, the information processing device 1 creates a list L of "uncorrected keywords" that are not to be corrected, and does not correct the keywords on list L.

[0014] On the other hand, in such cases, there are instances where keywords that should be corrected cannot be corrected. Specifically, as shown in Figure 1, when a search query containing multiple keywords such as "Kitami University" is entered from the user terminal 50 (step S11), the information processing device 1 may not be able to correct the search query that should have been corrected to "Kitami University" because "Kitami" and "University" are both included in list L.

[0015] Therefore, when a plurality of keywords included in a search query are included in the list L, the information processing apparatus 1 according to the embodiment estimates the degree of relevance between the keywords (step S12). The degree of relevance here is the co-occurrence degree or the mutual information amount. For example, the information processing apparatus 1 estimates that the higher the frequency with which a plurality of target keywords are used together based on the usage history of the search query, the higher the degree of relevance between the keywords.

[0016] That is, the information processing apparatus 1 estimates that the higher the search frequency for "Kitami University", the higher the degree of relevance between "Kitami" and "University". Then, when the estimated degree of relevance exceeds the threshold, the information processing apparatus 1 does not correct the search query, and when the estimated degree of relevance is below the threshold, the information processing apparatus 1 inputs the search query into a correction model to correct the search query (step S13).

[0017] After that, the information processing apparatus 1 executes a search using the search query that was not a correction target or the corrected search query (step S14), and provides the search result to the user terminal 50 (step S15).

[0018] In this way, the information processing apparatus 1 according to the embodiment pre-creates a list L of non-correction target keywords, and when the search query is composed of a plurality of keywords included in the list L, estimates the degree of relevance between the keywords.

[0019] Then, when the estimated degree of relevance between the keywords by the information processing apparatus 1 according to the embodiment exceeds the threshold, the search query is not corrected, and when the estimated degree of relevance between the keywords is below the threshold, the search query is corrected.

[0020] That is, the information processing apparatus 1 according to the embodiment can suppress overcorrection by using the list L, and can correct the keywords included in the list L by correcting the search query according to the degree of relevance between the keywords.

[0021] Therefore, according to the information processing device 1 of the embodiment, it is possible to correct to an appropriate search query while suppressing excessive correction.

[0022] [2. Information Processing Devices] Next, an example of the configuration of the information processing device 1 according to the embodiment will be described using Figure 2. Figure 2 is a block diagram showing an example of the configuration of the information processing device 1 according to the embodiment. As shown in Figure 2, the information processing device 1 comprises a communication unit 2, a storage unit 3, and a control unit 4. The information processing device 1 may also have an input unit (for example, a keyboard or mouse) that accepts various operations from an administrator or other user of the information processing device 1, and a display unit (for example, a liquid crystal display) for displaying various information.

[0023] The communication unit 2 is implemented, for example, by a NIC (Network Interface Card). The communication unit 2 is connected to a communication network such as 4G (4th Generation) or 5G (5th Generation) by wire or wireless connection, and transmits and receives information to and from each of the user terminals 50, etc., via the communication network.

[0024] The storage unit 3 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or by storage devices such as hard disks and optical discs. The storage unit 3 comprises a correction model storage unit 31, a list information storage unit 32, a search log storage unit 33, and a content storage unit 34.

[0025] The correction model storage unit 31 stores the correction model. The correction model is a machine learning model that uses the keyword to be corrected and the correct keyword for that keyword as ground truth data.

[0026] The list information storage unit 32 stores list information. The list information includes a list L (see Figure 1) of keywords to be left uncorrected. Examples of keywords registered in list L include place names. Although list L is registered by the administrator of the information processing device 1, it may also be automatically generated by the information processing device 1.

[0027] The search log storage unit 33 stores the search log. The search log is the search history of the search queries used by each user. Figure 3 is a diagram showing an example of the search log storage unit 33 according to this embodiment. As shown in Figure 3, the search log storage unit 33 stores information such as "search date and time," "user information," "location information," and "keywords" in association with each other.

[0028] The "Search Date and Time" field stores the date and time the search was performed. The "User Information" field stores information about the user who performed the corresponding search. Note that the user information may be, for example, a user ID, or information indicating various attributes of the user.

[0029] The "Location Information" field stores the location information of the user who performed the search. The "Keywords" field stores the keywords included in the search query entered by the user who performed the search.

[0030] If a search query contains multiple keywords, the "Keywords" field will store each keyword separately.

[0031] Returning to the explanation of Figure 2, let's describe the content storage unit 34. The content storage unit 34 stores various types of content. This content is the content that is searched in the search service provided by the information processing device 1.

[0032] The control unit 4 is a controller, and is realized by, for example, a CPU (Central Processing Unit) or MPU (Micro Processing Unit) executing various programs (corresponding to an example of an information processing program) stored in the memory device inside the information processing device 1, using RAM as the working area. Alternatively, the control unit 4 is a controller and can be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

[0033] As shown in Figure 2, the control unit 4 comprises a reception unit 41, a determination unit 42, an estimation unit 43, a correction unit 44, a search unit 45, and a provision unit 46, and realizes or executes the information processing functions and operations described below. Note that the internal configuration of the control unit 4 is not limited to the configuration shown in Figure 2, and other configurations are also possible as long as they perform the information processing described later. Also, the connection relationships of the various processing units in the control unit 4 are not limited to the connection relationships shown in Figure 2, and other connection relationships are also possible.

[0034] The reception unit 41 receives a search request, including the search query, from the user terminal 50. The reception unit 41 registers the search query received from the user terminal 50 in the search log storage unit 33 and passes it to the determination unit 42. The reception unit 41 may also receive user information and location information from the user terminal 50 when receiving the search request. The user information and location information correspond to an example of user context.

[0035] The determination unit 42 determines whether the keyword included in the search query received by the reception unit 41 is registered in list L stored in list information storage unit 32. If the search query is a single keyword and that keyword is registered in list L, the determination unit 42 passes the keyword to the search unit 45. In other words, in this case, the search query will be used for the search without correction. If the search query is a single keyword and that keyword is not registered in list L, it passes it to the correction unit 44.

[0036] Furthermore, if the search query contains multiple keywords, the determination unit 42 determines that one keyword is registered in list L and the other keyword is not registered in list L. In this case, it passes the keyword not registered in list L to the correction unit 44 and the keyword registered in list L to the search unit 45. In other words, in this case, the keyword registered in list L is not corrected, while the keyword not registered in list L is corrected.

[0037] Furthermore, if the determination unit 42 determines that a search query contains multiple keywords and that each of these keywords is registered in list L, it passes these keywords to the estimation unit 43. In other words, when multiple keywords are each registered in list L, the estimation unit 43 estimates the degree of relevance between the keywords.

[0038] The estimation unit 43 estimates the degree of relevance between keywords when it receives a search query that includes multiple keywords pre-set as non-correction targets. For example, the estimation unit 43 estimates the degree of relevance between keywords based on the frequency with which each keyword was used together in previously used search queries.

[0039] The estimation unit 43 estimates that the more frequently each keyword is used together, the higher the degree of relevance between the keywords. Alternatively, the estimation unit 43 may estimate the degree of relevance between keywords based on the frequency of occurrence of each keyword within the same content. In this case, the estimation unit 43 estimates that the more frequently each keyword is used in the same content, the higher the degree of relevance between the keywords. Note that the content here is a web document, but it may be any content.

[0040] Here, the estimation process of the estimation unit 43 will be explained using Figure 4. Figure 4 is a schematic diagram of the estimation process according to the embodiment. As shown in Figure 4, for example, suppose the search query is "Kitami University".

[0041] In this case, keyword 1 is "Kitami" and keyword 2 is "university". The estimation unit 43 estimates the degree of relevance between keyword 1 and keyword 2. If the estimated degree of relevance is low, it corrects the "Kitami University" search query using the correction model and obtains the corrected search query.

[0042] In other words, if the relevance between keyword 1 and keyword 2 is sufficiently high, the search query will be performed as is without correction; however, if the relevance between keyword 1 and keyword 2 is low, the search query will be corrected.

[0043] In this case, the estimation unit 43 may estimate the degree of relevance between each keyword according to the user context. The user context includes the user's location information, search history, user behavior history, etc.

[0044] For example, the estimation unit 43 estimates the degree of relevance between keywords based on the history of search queries by users in similar contexts. For example, suppose the search query to be estimated is "XXX YYY", and for one group of users, the search query "XXX YYY" is the correct search query, while for another group of users, the correct search query is "XXX ZZZ".

[0045] In this case, if the searcher's context is similar to that of a particular user group, the relevance of "XXX" and "ZZZ" is estimated to be high. Conversely, if the searcher's context is similar to that of a user group that accurately identifies with "XXX ZZZ," the relevance of "XXX" and "ZZZ" is estimated to be low.

[0046] In other words, the estimation unit 43 can estimate the degree of relevance between keywords according to the user context, thereby tailoring the estimation to the user. This allows for appropriate correction of the search query.

[0047] Furthermore, when estimating relevance based on location information, the estimation unit 43 divides the user group into categories based on location information and estimates relevance based on the search history within each category. This makes it possible to estimate relevance based on the user's location information.

[0048] Returning to the explanation of Figure 2, let's describe the correction unit 44. The correction unit 44 does not correct the search query if the relevance estimated by the estimation unit 43 exceeds a threshold, and corrects the search query if the relevance falls below the threshold. The correction unit 44 corrects the search query by inputting the search query to the correction model stored in the correction model storage unit 31.

[0049] When the correction unit 44 corrects a search query that contains multiple keywords registered in the list L that are not to be corrected, it converts the keywords to other keywords. For example, the correction unit 44 corrects the search query "Kitami University" to "Kitami University". In this case, the correction unit 44 may continue correcting the original search query until the relevance between the keywords in the corrected search query exceeds a threshold, or until it exceeds the relevance between the keywords before correction.

[0050] The search unit 45 is a so-called search engine, and it searches the content stored in the content storage unit 34 using a search query. In this case, the search unit 45 performs the search using the search query after correction processing by the correction unit 44. This allows the search unit 45 to perform the search with an appropriate search query.

[0051] The provisioning unit 46 provides the search results from the search unit 45 to the user terminal 50 via the communication unit 2.

[0052] [3. Processing Flow] Next, the processing procedure of the information processing device 1 according to the embodiment will be described using Figure 5. Figure 5 is a flowchart showing an example of the correction process according to the embodiment.

[0053] As shown in Figure 5, the information processing device 1 accepts a search query containing two or more keywords (step S111). Next, the information processing device 1 determines whether or not both keywords included in the search query are registered in list L (step S112).

[0054] If it is determined that both keywords are registered in list L (step S112; Yes), the information processing device 1 proceeds to step S113. If it is determined that neither keyword is registered in list L (step S112; No), it proceeds to step S115.

[0055] Next, the information processing device 1 estimates the degree of relevance between keywords included in the search query (step S113), and determines whether the estimated degree of relevance is less than a threshold (step S114).

[0056] If the information processing device 1 determines that the estimated relevance is less than the threshold (step S114; Yes), it corrects the search query (step S115). If it determines that the estimated relevance is greater than the threshold (step S114; No), it proceeds to step S116 without going through step S115.

[0057] Next, the information processing device 1 performs a search using the uncorrected or corrected search query (step S116) and provides the search results (step S117). Then, the information processing device 1 terminates the process.

[0058] [4. Variations] In the embodiments described above, we explained a case in which the necessity of correction is determined based on the degree of relevance between keywords when correcting errors or other mistakes included in a search query, but the invention is not limited to this. For example, the present invention may also be applied to determining the necessity of correction based on the degree of relevance between keywords for any content, including text.

[0059] [5. Effects] The information processing device 1 according to this embodiment includes an estimation unit 43 that estimates the degree of relevance between keywords when it receives a search query containing a plurality of keywords that have been set in advance as not to be corrected, and a correction unit 44 that does not correct the search query if the degree of relevance estimated by the estimation unit 43 exceeds a threshold, and corrects the search query if the degree of relevance falls below the threshold.

[0060] Furthermore, the estimation unit 43 estimates the degree of relevance between keywords based on the frequency with which each keyword has been used together in past search queries. Additionally, the estimation unit 43 estimates the degree of relevance between keywords based on the frequency with which each keyword appears in the same content.

[0061] Furthermore, when correcting a search query, the correction unit 44 corrects keywords that should not be corrected to other keywords. In addition, the estimation unit 43 estimates the degree of relevance between each keyword based on the user context of the user who entered the search query.

[0062] Furthermore, the information processing device 1 includes a providing unit 46 that provides search results obtained using the search query after correction processing by the correction unit 44.

[0063] By any or a combination of the above-described processes, the information processing device according to this application can correct search queries to appropriate ones while suppressing excessive corrections.

[0064] [6. Hardware Configuration] Furthermore, the information processing device 1 according to the above embodiment is realized by a computer 1000 having a configuration such as that shown in Figure 6. Figure 6 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device according to the embodiment. The computer 1000 has a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

[0065] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, controlling various components. The ROM 1300 stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0066] The HDD1400 stores programs executed by the CPU1100, as well as data used by such programs. The communication interface1500 receives data from other devices via the network (communication network) N and sends it to the CPU1100, and transmits data generated by the CPU1100 to other devices via the network N.

[0067] The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice (in Figure 6, output devices and input devices are collectively referred to as "input / output devices") via the input / output interface 1600. The CPU 1100 acquires data from input devices via the input / output interface 1600. The CPU 1100 also outputs the generated data to output devices via the input / output interface 1600.

[0068] The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0069] For example, when the computer 1000 functions as an information processing device according to the embodiment, the CPU 1100 of the computer 1000 realizes the functions of the control unit 4 by executing programs loaded on the RAM 1200. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be obtained from other devices via the network N.

[0070] [7. Other] Although embodiments of the present invention have been described above, the present invention is not limited by the content of these embodiments. Furthermore, the aforementioned components include those that can be easily conceived by those skilled in the art, those that are substantially the same, and those that fall within the so-called equivalent range. Moreover, the aforementioned components can be combined as appropriate. Furthermore, various omissions, substitutions, or modifications of the components can be made without departing from the gist of the embodiments described above.

[0071] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0072] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0073] For example, the information processing device described above may be implemented using multiple server computers, and depending on the function, it may be implemented by calling external platforms, etc., via APIs (Application Programming Interfaces) or network computing, allowing for flexible configuration changes.

[0074] Furthermore, the embodiments and modifications described above can be combined as appropriate, provided that the processing content is not inconsistent.

[0075] Furthermore, the terms "section, module, unit" mentioned above can be replaced with "means" or "circuit," etc. For example, the acquisition unit can be replaced with acquisition means or acquisition circuit. [Explanation of Symbols]

[0076] 1. Information Processing Device 2 Communications Department 3 Storage section 4. Control Unit 31 Correction Model Memory Unit 32 List Information Storage Unit 33 Search log storage unit 34 Content Storage Unit 41 Reception Department 42 Judgment section 43 Estimation part 44 Correction Section 45 Search Section 46 Providing Department 50 User terminals

Claims

1. When a search query is received that includes multiple keywords pre-set as not to be corrected, an estimation unit estimates the degree of relevance between each keyword, If the degree of relevance estimated by the estimation unit exceeds a threshold, the search query is not corrected; if the degree of relevance falls below the threshold, the correction unit corrects the search query. An information processing device characterized by comprising:

2. The estimation unit, To estimate the degree of relevance between each keyword based on the frequency with which each keyword was used together in past search queries. The information processing apparatus according to claim 1, characterized by the following:

3. The estimation unit, To estimate the degree of relevance between each keyword based on the frequency of occurrence of each keyword in the same content. The information processing apparatus according to claim 1, characterized by the following:

4. The correction section is, When correcting the aforementioned search query, the keyword that is not to be corrected is corrected to another keyword. The information processing apparatus according to claim 1, characterized by the following:

5. The estimation unit, Based on the user context of the user who entered the search query, the degree of relevance between each of the keywords is estimated. The information processing apparatus according to claim 1, characterized by the following:

6. A providing unit that provides search results obtained using the search query after correction processing by the correction unit. To be prepared The information processing apparatus according to claim 1, characterized by the following:

7. A method of information processing performed by a computer, When a search query containing multiple keywords pre-configured as non-correction targets is received, the estimation process involves estimating the relevance between each keyword, If the relevance estimated by the estimation step exceeds a threshold, the search query is not corrected; if the relevance falls below the threshold, the search query is corrected in the correction step. An information processing method characterized by including

8. When a search query containing multiple keywords pre-configured as non-correction targets is received, an estimation procedure is provided to estimate the relevance between each keyword, If the relevance estimated by the estimation procedure exceeds a threshold, the search query is not corrected; if the relevance falls below the threshold, the search query is corrected using the correction procedure. An information processing program characterized by causing a computer to execute it.

Images