Information processing apparatus, information processing method, and information processing program

The information processing device efficiently classifies user content by extracting and evaluating keyword combinations based on part-of-speech patterns, addressing inefficiencies in conventional methods and enhancing accuracy through vector-based clustering.

JP2024166719APending Publication Date: 2024-11-29LY CORP

Patent Information

Application Number
JP2023083022
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-19
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

Conventional methods for classifying user-posted content are inefficient and prone to inaccuracies due to high processing loads and improper vectorization of text content.

Method used

An information processing device that extracts a predetermined number of keywords from user content, evaluates their combinations based on part-of-speech patterns, and converts them into vectors for accurate classification using clustering.

Benefits of technology

Enables efficient and accurate classification of user content by focusing on keyword combinations that mimic user search queries, reducing processing loads and improving classification accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024166719000001_ABST
    Figure 2024166719000001_ABST
Patent Text Reader

Abstract

To provide an information processing apparatus, an information processing method, and an information processing program for efficiently classifying a posted content.SOLUTION: In an information processing apparatus which is implemented by a cloud system and a server and manages various media sites, such as a news site, a control unit includes an extraction unit, an evaluation unit, a conversion unit, and a classifying unit. The extraction unit extracts a predetermined number of keywords from posted contents posted in a predetermined period (S2), out of posted contents posted by a user U (S1). The evaluation unit specifies a part of speech of each keyword, and compares combinations of the specified parts of speech with a part-of-speech pattern registered on a part-of-speech pattern dictionary (S3). The conversion unit converts combinations of the keywords extracted by the extraction unit, for each of the posted content, to vectors (S4). The classifying unit performs clustering based on similarities of the vectors converted by the conversion unit and classifies the posted content (S5).SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

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

[0002] 2. Description of the Related Art Conventionally, there are services that provide various types of content. For example, such services may provide various types of content posted by users on microblogs, social networking services (SNS), and the like. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2022-47013 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the conventional techniques leave room for improvement in terms of efficiently classifying the posted content posted by users.

[0005] The present invention has been made in consideration of the above, and has an object to provide an information processing device, an information processing method, and an information processing program that are capable of efficiently classifying posted content. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the objective, the information processing device of the present invention includes an extraction unit that extracts a predetermined number of keywords from posted content posted by users within a predetermined period of time, a conversion unit that converts the combination of keywords extracted by the extraction unit into vectors, and a classification unit that classifies the posted content into each category based on the similarity of the vectors converted by the conversion unit. Effect of the Invention

[0007] According to the present invention, posted content can be efficiently classified. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of content according to the embodiment. [Diagram 3] FIG. 3 is a block diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of information stored in a content storage unit according to the embodiment. [Diagram 5] FIG. 5 is a flowchart illustrating an example of a processing procedure of the classification processing according to the embodiment. [Figure 6] FIG. 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. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, the information processing device, the information processing method, and the information processing program according to the present application will be described in detail with reference to the drawings. Note that the information processing device, the information processing method, and the information processing program according to the present application are not limited to the embodiments.

[0010] [Embodiment] [1. Information Processing] First, an example of information processing according to the embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of information processing according to the embodiment. Note that the information processing according to the embodiment is realized by an information processing device 1 shown in Fig. 1.

[0011] 1 is an information processing device that operates various media sites such as a news site, etc. The information processing device 1 is realized by, for example, a cloud system or a server.

[0012] The user terminal 100 shown in Fig. 1 is a terminal device owned by a user U. For example, the user U operates the user terminal 100 to provide news articles and the like to the user U through various media sites. Note that, although Fig. 1 illustrates an example in which the user terminal 100 is a smartphone, the user terminal 100 may be other devices such as a PC (Personal Computer).

[0013] 1 is a server device that manages various SNSs (Social Networking Services). For example, some articles (contents) of the SNSs managed by the SNS server 200 are posted on various media sites managed by the information processing device 1.

[0014] Since each user posts content on the SNS at any time, the content is timely. Therefore, the information processing device 1 can post timely articles by posting content using the SNS on a news site or the like.

[0015] On the other hand, when posting articles about submitted content on news sites, etc., it is necessary to classify each submitted content by topic. For example, one classification method is to use a method such as Word2Vec to vectorize the submitted content and cluster it according to the distance between the vectors.

[0016] However, in the above-mentioned method, all text contained in the posted content is vectorized, which may increase the processing load required for vectorization, and depending on the accuracy of the vectorization, the posted content may not be converted into an appropriate vector, and may not be classified appropriately.

[0017] Therefore, in this embodiment, keywords are extracted from the posted content, and the extracted keywords are converted into vectors. Furthermore, in this embodiment, the posted content can be classified with high accuracy by evaluating the combinations of the extracted keywords and then vectorizing them.

[0018] 1, the information processing device 1 acquires posted content from the SNS server 200 (step S1). For example, the information processing device 1 acquires posted content posted to the SNS server 200 within a predetermined time. For example, the information processing device 1 acquires posted content posted to the SNS in real time at a predetermined cycle, and provides the user terminal 100 with, for example, content shown in FIG. 2 through processing described later.

[0019] Next, the information processing device 1 extracts a predetermined number of keywords from each of the posted contents (step S2). For example, the keywords extracted by the information processing device 1 are characteristic words contained in the posted contents. The example shown in Fig. 1 shows a case where keywords such as "Tokyo Bay", "fireworks display", "live broadcast", and "three-foot fireworks ball" are extracted from one posted content.

[0020] Next, the information processing device 1 randomly combines the extracted keywords and evaluates each combination of keywords (step S3). For example, the information processing device 1 evaluates the combination of parts of speech of the keywords included in each combination. Here, when keywords are randomly combined, depending on the combination, it may be difficult to interpret the content of the original posted content.

[0021] In this embodiment, the focus is on search queries used by users when performing searches, and each keyword combination is evaluated by imitating the search queries. Specifically, when a user searches by combining multiple keywords, each keyword combination can be regarded as information that summarizes the contents that the user wants.

[0022] That is, a keyword combination that has a similar tendency to a multiple-word search query by a user is likely to be a keyword combination that succinctly expresses the content of the posted content. Therefore, the information processing device 1 evaluates each keyword combination by creating a dictionary of the tendency of multiple-word search queries used by users and comparing it with each keyword combination.

[0023] More specifically, the keyword combinations are evaluated using a part-of-speech pattern dictionary created based on the part-of-speech combinations in three-word search queries used by users. The part-of-speech pattern dictionary is a dictionary of data that contains tendencies in part-of-speech patterns in three-word search queries used by users.

[0024] As an example, if the part-of-speech pattern of a three-word search query tends to be "proper noun, common noun, sa-variant noun," the part-of-speech pattern dictionary is configured to highly evaluate keyword combinations consisting of "proper noun, common noun, sa-variant noun."

[0025] Next, when the information processing device 1 finishes evaluating each keyword combination, it converts the most highly evaluated keyword combination into a vector (step S3). In Fig. 1, the keyword combination Sc1 consisting of "F1, F2, F3" has the highest evaluation, and the combination is converted into a vector V1.

[0026] As a result, the information processing device 1 vectorizes only the keyword combinations that capture the details of the posted content. Note that the vectorization of the keyword combinations is performed by converting each keyword into a vector using any method such as Word2Vec and calculating the sum of these vectors.

[0027] Then, the information processing device 1 performs the processes of steps S2 to S4 on each piece of posted content acquired in step S1. Then, the information processing device 1 performs clustering (classification) using vectors attributable to each piece of posted content (step S5). Note that the clustering is performed by classifying each vector based on the distance between each vector.

[0028] 1 shows an example in which the posted content corresponding to vector V1 is classified into a category (topic) of fireworks festivals, and the posted content corresponding to vector V20 is classified into a category of Ukraine. Note that, as will be described later, the information processing device 1 can also generate a category name for each category.

[0029] In this way, the information processing device 1 can appropriately classify each posted content by vectorizing the combination of keywords and clustering them. After that, the information processing device 1 generates content and provides it to the user terminal 100 (step S6).

[0030] Fig. 2 is a diagram showing an example of content provided by the information processing device 1. As shown in Fig. 2, the content is a word cloud including keywords for each category (corresponding to "anime / manga", "IT", "idols", etc. shown in Fig. 2).

[0031] The information processing device 1 extracts multiple keywords from each of the topics classified by clustering, and provides a word cloud including the extracted keywords. In the example shown in Fig. 2, each of the clusters such as "Turkey, Syria, earthquake" and "society, same-sex marriage, criticism" indicates that they are keywords extracted from one topic.

[0032] In addition, in the word cloud, the display area and font size are determined according to the popularity of each topic. The popularity here is, for example, the number of posted contents in each topic. In other words, the more posts a category has, the larger the display area and font size will be. Note that the font color is actually changed according to the popularity, and a more highlighted color is used for a topic with a higher popularity.

[0033] Furthermore, when the user selects a keyword displayed on the word cloud, the information processing device 1 provides the user with posted content that belongs to the topic of the selected keyword.

[0034] In this way, by displaying multiple keywords for one topic on the word cloud, users can easily understand the content of each topic.

[0035] [2. Information Processing Device] Next, a configuration example of the information processing device 1 according to the embodiment will be described with reference to Fig. 3. Fig. 3 is a block diagram showing a configuration example of the information processing device 1 according to the embodiment. As shown in Fig. 3, the information processing device 1 includes a communication unit 2, a storage unit 3, and a control unit 4. Note that the information processing device 1 may also include an input unit (e.g., a keyboard, a mouse, etc.) that receives various operations from an administrator who uses the information processing device 1, and a display unit (e.g., a liquid crystal display, etc.) that displays various information.

[0036] The communication unit 2 is realized by, for example, a network interface card (NIC) etc. The communication unit 2 is connected to a communication network such as 4G (4th Generation) or 5G (5th Generation) by wire or wirelessly, and transmits and receives information to and from each of the user terminal 100, the SNS server 200, etc. via the communication network.

[0037] The storage unit 3 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 3 has a content storage unit 31 and a dictionary information storage unit 32.

[0038] The content storage unit 31 stores posted content. The posted content is content posted by each user to the SNS server 200. Fig. 4 is a diagram showing an example of information stored in the content storage unit 31 according to the embodiment.

[0039] As shown in FIG. 4, the content storage unit 31 stores information in the fields of "ID", "main body", "keyword group", "vector", "category", etc. in association with each other. The "ID" field stores an identifier for identifying each piece of posted content. The "main body" field stores data on the posted content itself identified by the corresponding ID. The main body includes text, images, videos, etc.

[0040] The "keyword group" item stores a keyword group included in the body of the posted content identified by the corresponding ID. For example, the keyword group is a characteristic word included in the posted content.

[0041] The "vector" field stores a vector indicating the posted content identified by the corresponding ID. As described above, the vector is a vector of the most highly rated combination of keywords extracted from the posted content.

[0042] The "Category" item stores the category of the posted content identified by the corresponding ID. The category is classified based on the corresponding vector using a classification method such as clustering.

[0043] Returning to the explanation of FIG. 3, the dictionary information storage unit 32 will be explained. The dictionary information storage unit 32 stores dictionary information. The dictionary information is information related to the part-of-speech pattern dictionary described above. The part-of-speech pattern dictionary is data in which part-of-speech patterns in a search query of multiple words used by a user are compiled into a dictionary. For example, the part-of-speech pattern dictionary is generated based on each search query acquired from a search server or the like.

[0044] Next, the control unit 4 will be described. The control unit 4 is a controller, and is realized, for example, by a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) executing various programs (corresponding to an example of an information processing program) stored in a storage device inside the information processing device 1 using a RAM as a working area. The control unit 4 is also, for example, a controller, and is realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0045] As shown in Fig. 3, the control unit 4 includes an acquisition unit 41, an extraction unit 42, an evaluation unit 43, a conversion unit 44, a classification unit 45, a generation unit 46, and a provision unit 47, and realizes or executes the functions and actions of information processing described below. Note that the internal configuration of the control unit 4 is not limited to the configuration shown in Fig. 3, and may be other configurations as long as they perform the information processing described below. Also, the connection relationship between the processing units of the control unit 4 is not limited to the connection relationship shown in Fig. 3, and may be other connection relationships.

[0046] The acquisition unit 41 acquires posted content from the SNS server 200. For example, the acquisition unit 41 acquires posted content posted within a period from the SNS server 200 for each period, and registers the acquired content in the content storage unit 31.

[0047] At this time, the acquisition unit 41 may acquire all posted contents, or may acquire only posted contents related to current trends in the SNS, for example. In this case, the acquisition unit 41 acquires posted contents including trending keywords and hashtags. In other words, the acquisition unit 41 may acquire contents that are attracting attention in real time.

[0048] The extraction unit 42 extracts a predetermined number of keywords from the posted content posted within a predetermined period of time among the posted content posted by the user. First, the extraction unit 42 extracts all characteristic words from the text of each content and registers them in a keyword group in the content storage unit 31. Note that the extraction of characteristic words can be performed, for example, by extracting keywords registered in a dictionary of characteristic words from the text of the posted content.

[0049] Next, the extraction unit 42 extracts a predetermined number of combinations (for example, three words) from the extracted keywords. For example, the extraction unit 42 extracts combinations of each keyword by combining the extracted keywords in a brute force manner.

[0050] At this time, the extraction unit 42 extracts combinations of each keyword without changing the order in which each keyword appears in the text. That is, the extraction unit 42 extracts combinations of each keyword while maintaining the order in which the keywords appear in the text. For example, from a text that includes four keywords in the order "A, B, C, D", the extraction unit 42 extracts combinations of four keywords, "A, B, C", "A, B, D", "A, C, D", and "B, C, D".

[0051] The evaluation unit 43 evaluates the keyword combinations extracted from each posted content by the extraction unit 42 based on the combinations of parts of speech. Specifically, the evaluation unit 43 evaluates each keyword combination by comparing the part of speech pattern in each keyword combination with the part of speech pattern dictionary stored in the dictionary information storage unit 32.

[0052] First, the evaluation unit 43 identifies the parts of speech of each keyword, and compares the identified combination of parts of speech with the part-of-speech patterns registered in the part-of-speech pattern dictionary. At this time, the evaluation unit 43 evaluates each combination of keywords taking into account the order of the parts of speech. The evaluation unit 43 may also perform evaluation by weighting each type of part of speech. For example, the evaluation unit 43 may evaluate proper nouns higher than other parts of speech. For example, the evaluation unit 43 may also evaluate each combination of keywords using multiple part-of-speech pattern dictionaries generated for each user attribute.

[0053] Then, the evaluation unit 43 passes the keyword combination with the highest score among the keyword combinations to the conversion unit 44. That is, the evaluation unit 43 calculates the goodness of style of the combination of parts of speech in three words as a score, and passes the keyword combination with the highest score to the subsequent processing. This makes it possible to extract a keyword combination that is close to the tendency of parts of speech used in a search query by a user, that is, a keyword combination that is likely to be used as a search query by a user.

[0054] If the evaluation unit 43 evaluates each keyword combination and finds that the highest score is equal to or less than a threshold, the evaluation unit 43 may stop subsequent processing of the corresponding posted content.

[0055] The conversion unit 44 converts the keyword combination extracted by the extraction unit 42 into a vector. The conversion unit 44 converts the keyword combination that has been most highly evaluated by the evaluation unit 43 into a vector. The conversion unit 44 converts each keyword into a vector using Word2Vec, Transformer, or the like, and the sum of these vectors represents the vector of the keyword combination.

[0056] The classification unit 45 classifies the posted content into each category based on the similarity of the vectors converted by the conversion unit 44. The classification unit 45 classifies the posted content corresponding to each vector by clustering each vector according to the distance. In other words, the classification unit 45 classifies vectors that are close to each other into the same category, and vectors that are far from each other into different categories. Note that the classification unit 45 may merge similar categories (for example, categories whose centers of gravity are close to each other) as a result of classification into each category.

[0057] For example, each category contains content related to a common topic. In this way, by vectorizing a combination of keywords extracted from the posted content, rather than the posted content itself, and then performing clustering, the posted content can be classified more easily and accurately than if the posted content were directly vectorized.

[0058] The generating unit 46 generates a category name for each category classified by the classifying unit 45. For example, the generating unit 46 converts the centroid (center of gravity) of each category into text and generates it as the category name of each category.

[0059] At this time, the generating unit 46 may set, for example, a plurality of words (for example, three words) close to the centroid of each category as the category name. Also, at this time, the generating unit 46 may evaluate the three words that will be the category name using a part-of-speech pattern dictionary, and if the score exceeds a threshold, the three words may be set as the category name.

[0060] This enables the generation unit 46 to generate category names that are combinations of parts of speech that are likely to be used as search queries when a user searches each category using multiple words.

[0061] The providing unit 47 provides various contents to the user terminal 100. For example, the providing unit 47 provides a word cloud such as that shown in Fig. 2 to the user terminal 100. For example, the word cloud is a text representation of the contents of each category that is attracting attention in real time.

[0062] The provision unit 47 sets the font size, color, size, etc. of the three-word category name of each category according to the popularity of each category, generates a word cloud arranged within an image, and provides (transmits) the word cloud to the user terminal 100 to which it is to be provided.

[0063] For example, the attention level corresponds to the number of posts for each topic, and the more posts there are, the higher the attention level. Note that the attention level may be adjusted according to the interests of the user to whom the information is provided. In other words, the attention level may be increased for categories in which the user is interested.

[0064] [3. Processing flow] Next, a processing procedure executed by the information processing device 1 according to the embodiment will be described with reference to Fig. 5. Fig. 5 is a flowchart showing an example of the processing procedure of the classification process according to the embodiment. Note that the processing shown below is repeatedly executed by the information processing device 1 at a predetermined cycle.

[0065] 5, the information processing device 1 first acquires posted contents posted during a predetermined period (step S101), and then extracts keywords from each of the posted contents (step S102).

[0066] Next, the information processing device 1 evaluates the extracted keyword combination using a part-of-speech pattern dictionary (step S103). Next, the information processing device 1 vectorizes the keyword combination with the highest score (step S104).

[0067] Next, the information processing device 1 classifies the posted contents by clustering the vectors (step S105).Then, the information processing device 1 generates a topic name for each category (step S106) and ends the process.

[0068] 4. Modifications In the above-described embodiment, the information processing device 1 classifies the posted content posted on the SNS, but the present invention is not limited to this. The posted content may be other content (e.g., a news article) as long as it contains text. The posted content may also be, for example, a reply to an arbitrary post. The posted content may also include text that is generated by transcribing audio data using AI or the like. In other words, audio and video posted to each medium may be regarded as posted content.

[0069] 5. Effects The information processing device 1 according to the embodiment includes an extraction unit 42 that extracts a predetermined number of keywords from each piece of posted content posted by a user within a predetermined period of time, a conversion unit 44 that converts the combination of keywords extracted by the extraction unit 42 into a vector for each piece of posted content, and a classification unit 45 that classifies the posted content into categories based on the similarity of the vectors converted by the conversion unit 44.

[0070] Moreover, the information processing device 1 according to the embodiment includes a generating unit 46 that generates a topic name for each category classified by the classifying unit 45. Moreover, the generating unit 46 generates a topic name based on the center of gravity of the category.

[0071] In addition, the information processing device 1 according to the embodiment includes an evaluation unit 43 that evaluates the keyword combinations extracted by the extraction unit 42 based on the combinations of parts of speech, and a conversion unit 44 converts into a vector a keyword combination for which the evaluation result by the evaluation unit 43 exceeds a predetermined threshold.

[0072] The evaluation unit 43 also evaluates the keyword combination based on the tendency of the combination of parts of speech in the multi-word search query used by the user. The evaluation unit 43 also compares the order of the parts of speech in the multi-word search query used by the user with the order of the parts of speech in the keyword combination to evaluate the keyword combination.

[0073] By performing any one or a combination of the above-described processes, the information processing device according to the present application can appropriately classify posted content.

[0074] [6. Hardware Configuration] The information processing device 1 according to the embodiment described above is realized by a computer 1000 having a configuration as shown in Fig. 6. Fig. 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, a RAM 1200, a ROM 1300, a HDD 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F) 1700.

[0075] The CPU 1100 operates and controls each unit based on a program stored in the ROM 1300 or the HDD 1400. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started up, programs that depend on the hardware of the computer 1000, and the like.

[0076] The HDD 1400 stores programs executed by the CPU 1100, data used by such programs, etc. The communication interface 1500 receives data from other devices via a network (communication network) N and sends the data to the CPU 1100, and transmits data generated by the CPU 1100 to other devices via the network N.

[0077] The CPU 1100 controls output devices such as a display and a printer, and input devices such as a keyboard and a mouse (in FIG. 6, the output devices and the input devices are collectively referred to as "input / output devices") via an input / output interface 1600. The CPU 1100 acquires data from the input devices via the input / output interface 1600. The CPU 1100 also outputs generated data to the output devices via the input / output interface 1600.

[0078] 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 a PD (Phase change rewritable Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0079] For example, when the computer 1000 functions as an information processing device according to the embodiment, the CPU 1100 of the computer 1000 executes programs loaded onto the RAM 1200 to realize the functions of the control unit 4. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, the CPU 1100 may obtain these programs from another device via the network N.

[0080] [7. Other] Although the embodiments of the present application have been described above, the present invention is not limited to the contents of these embodiments. The above-described components include those that can be easily imagined by a person skilled in the art, those that are substantially the same, and those that are within the so-called equivalent range. Furthermore, the above-described components can be appropriately combined. Furthermore, various omissions, substitutions, or modifications of the components can be made without departing from the spirit of the above-described embodiments.

[0081] 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 a known method. In addition, the information including the processing procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

[0082] In addition, each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads, usage conditions, etc.

[0083] For example, the above-mentioned information processing device may be realized using multiple server computers, and depending on the functions, the configuration can be flexibly changed, such as by calling an external platform using an API (Application Programming Interface) or network computing.

[0084] Furthermore, the above-described embodiments and modifications can be appropriately combined as long as the processing contents are not contradictory.

[0085] Furthermore, the above-mentioned "section, module, unit" can be read as "means" or "circuit", etc. For example, an acquisition section can be read as an acquisition means or an acquisition circuit. [Explanation of symbols]

[0086] 1. Information processing device 2. Communications Department 3 Storage section 4. Control section 31 Content storage unit 32 Dictionary information storage unit 41 Acquisition Department 42 Extraction part 43 Evaluation Department 44 Conversion unit 45 Classification Department 46 Generation part 47 Providing Department 100 user terminals

Claims

1. an extracting unit that extracts a predetermined number of keywords from each piece of content posted by a user within a predetermined period of time; a conversion unit that converts the combination of the keywords extracted by the extraction unit into a vector for each piece of posted content; a classification unit that classifies the posted content based on the similarity of the vector converted by the conversion unit; An information processing device comprising:

2. a generation unit that generates a topic name for each category classified by the classification unit based on the vector; The information processing device according to claim 1 .

3. The generation unit is generating the topic names based on the centroids of the vectors in categories; 3. The information processing apparatus according to claim 2,

4. an evaluation unit that evaluates the combination of the keywords extracted by the extraction unit based on a combination of parts of speech; Equipped with The classification unit is classifying the posted content by using the vector of the combination of the keywords for which the evaluation by the evaluation unit exceeds a predetermined threshold value; 2. The information processing apparatus according to claim 1,

5. The evaluation unit is Evaluating the keyword combinations based on trends in part-of-speech combinations in multi-word search queries used by users.

5. The information processing apparatus according to claim 4,

6. The evaluation unit is evaluating the keyword combination by comparing the order of parts of speech in a multi-word search query used by a user with the order of parts of speech in the keyword combination; 6. The information processing device according to claim 5,

7. The conversion unit is extracting the combination of the keywords that is most highly evaluated by the evaluation unit and converting it into the vector; The classification unit is classifying the posted content by using the vector of the combination of the keywords extracted by the conversion unit; 5. The information processing apparatus according to claim 4,

8. 1. A computer-implemented information processing method, comprising: an extraction step of extracting a predetermined number of keywords from each of the posted contents posted by a user within a predetermined period of time; a conversion step of converting the combination of keywords extracted in the extraction step into a vector for each piece of posted content; a classification step of classifying the posted content based on the similarity of the vectors converted by the conversion step; 13. An information processing method comprising:

9. An extraction step of extracting a predetermined number of keywords from each of the posted contents posted by a user within a predetermined period of time; a conversion step of converting the combination of keywords extracted by the extraction step into a vector for each piece of posted content; a classification step of classifying the posted content based on a similarity of the vector converted by the conversion step; An information processing program characterized by causing a computer to execute the above.

Citation Information

Patent Citations

  • Mass analysis system and method for correcting conversion formula

    JP2022047013A

Cited By

  • Report generation apparatus and method

    WO2026135163A1