Information processing apparatus

Abstract

To provide an information processing apparatus for evaluating content that further closely matches user's preferences.SOLUTION: In an information processing apparatus that comprises a control unit, a communication unit, and a storage unit, the control unit comprises: extracting text information about a store from content of the store (S1); generating a content value vector in a plurality of dimensions from the text information (S2); calculating a person parameter of the user from user input information (S3); generating a user preference value vector in the plurality of dimensions from the person parameter (S4); and determining a similarity between the content value vector and the user preference value vector (S5).SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus.

Background Art

[0002] Conventionally, there is known a technique for presenting information on restaurants whose degree of coincidence between the taste tendency information of restaurants obtained from a restaurant database and the preference information of users obtained by questionnaires is equal to or higher than a predetermined value (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the above background art, restaurants are evaluated only based on evaluations related to taste. For this reason, other information about taste may not match the preferences of users, and it may not be possible to obtain user satisfaction.

[0005] An object of the present disclosure made in view of such circumstances is to evaluate content that more conforms to the preferences of users.

Means for Solving the Problems

[0006] An information processing apparatus according to an embodiment of the present disclosure is an information processing apparatus including a control unit, where the control unit extracts text information related to the store from the content of the store, generates a content value vector in multiple dimensions from the text information, calculates the personal parameters of the user from the input information of the user, generates a user preference value vector in the multiple dimensions from the personal parameters. Determining the degree of similarity between the content value vector and the user preference value vector; Performing an operation including this.

Effect of the Invention

[0007] According to an embodiment of the present disclosure, it is possible to evaluate content that more conforms to the preferences of the user.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Mode for Carrying Out the Invention

[0009] FIG. 1 is a schematic diagram of an information processing system S according to the present embodiment. The information processing system S includes an information processing apparatus 1 and a user terminal 2 that can communicate with each other via a network NW. The network NW includes, for example, a mobile communication network, a fixed communication network, or the Internet.

[0010] In FIG. 1, for simplicity of explanation, one information processing apparatus 1 and one user terminal 2 are shown. However, the number of each of the information processing apparatus 1 and the user terminal 2 is not limited to this. For example, the processing executed by the information processing apparatus 1 of the present embodiment may be executed by a plurality of information processing apparatuses 1 arranged in a distributed manner.

[0011] The information processing apparatus 1 is installed in a facility such as a data center. The information processing apparatus 1 is a computer such as a server belonging to a cloud computing system or other computing system.

[0012] The internal configuration of the information processing apparatus 1 will be described in detail with reference to FIG. 2.

[0013] The information processing apparatus 1 includes a control unit 11, a communication unit 12, and a storage unit 13. Each component of the information processing apparatus 1 is communicably connected to each other via, for example, a dedicated line.

[0014] The control unit 11 includes, for example, one or more general-purpose processors including a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). The control unit 11 may include one or more dedicated processors specialized for specific processing. Instead of including a processor, the control unit 11 may include one or more dedicated circuits. The dedicated circuit may be, for example, an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). The control unit 11 may include an ECU (Electronic Control Unit). The control unit 11 transmits and receives arbitrary information via the communication unit 12.

[0015] The communication unit 12 includes a communication module that conforms to one or more wired or wireless LAN (Local Area Network) standards for connecting to the network NW. The communication unit 12 may include a module that conforms to one or more mobile communication standards including LTE (Long Term Evolution), 4G (4th Generation), or 5G (5th Generation). The communication unit 12 may include a communication module or the like that conforms to one or more short-range communication standards or specifications including Bluetooth (registered trademark), AirDrop (registered trademark), IrDA, ZigBee (registered trademark), Felica (registered trademark), or RFID. The communication unit 12 transmits and receives any information via the network NW.

[0016] The storage unit 13 includes, for example, a semiconductor memory, a magnetic memory, an optical memory, or a combination of at least two of these, but is not limited thereto. The semiconductor memory is, for example, a RAM or a ROM. The RAM is, for example, an SRAM or a DRAM. The ROM is, for example, an EEPROM. The storage unit 13 may function as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 13 may store the information of the result analyzed or processed by the control unit 11. The storage unit 13 may store various information related to the operation or control of the information processing apparatus 1. The storage unit 13 may store a system program, an application program, and embedded software, etc. The storage unit 13 may be provided outside the information processing apparatus 1 and accessed from the information processing apparatus 1. The storage unit 13 includes a similarity DB.

[0017] The user terminal 2 is, for example, a mobile device such as a mobile phone, a smartphone, a wearable device, or a tablet. As an alternative example, the user terminal 2 may be a general-purpose device such as a PC or a dedicated device. "PC" is an abbreviation for personal computer.

[0018] The internal configuration of the user terminal 2 will be described in detail with reference to FIG. 3.

[0019] The user terminal 2 includes a control unit 21, a communication unit 22, a storage unit 23, a display unit 24, an input unit 25, and an imaging unit 26. Each component of the user terminal 2 is communicably connected to each other via, for example, a dedicated line.

[0020] The description of the hardware configurations of the control unit 21, the communication unit 22, and the storage unit 23 of the user terminal 2 may be the same as the description of the hardware configurations of the control unit 11, the communication unit 12, and the storage unit 13 of the information processing apparatus 1. The description here is omitted.

[0021] The display unit 24 is, for example, a display. The display is, for example, an LCD or an organic EL display. "LCD" is an abbreviation for liquid crystal display. "EL" is an abbreviation for electro luminescence. Instead of being provided in the user terminal 2, the display unit 24 may be connected to the user terminal 2 as an external output device. As a connection method, for example, any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used.

[0022] The input unit 25 is, for example, a physical key, a capacitive key, a pointing device, a touch screen provided integrally with the display, or a microphone. The input unit 25 receives an operation for inputting information used for the operation of the user terminal 2. Instead of being provided in the user terminal 2, the input unit 25 may be connected to the user terminal 2 as an external input device. As a connection method, for example, any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used. "USB" is an abbreviation for Universal Serial Bus. "HDMI (registered trademark)" is an abbreviation for High-Definition Multimedia Interface.

[0023] The imaging unit 26 includes a camera. The imaging unit 26 can image the surroundings. The imaging unit 26 may record the captured image in the storage unit 23 or transmit it to the control unit 21 for image analysis. The image includes a still image or a moving image.

[0024] The information processing method executed in the information processing system S will be described in detail below.

[0025] The control unit 11 of the information processing apparatus 1 extracts text information related to a store from the content of an arbitrary or specific store by scraping, purchasing a database, or collecting through a Web API. The store may be, for example, a massage parlor, a karaoke parlor, or a restaurant. The content may include, for example, tags, descriptions, or reviews. The tags may be, for example, "congee", "calorie", "kebab", "stand-up eating", "French". The descriptions may be, for example, "a fashionable store", "attentive to customers", "ingredients with special attention". The reviews may be, for example, "crowded", "okay for one person", "delicious".

[0026] The control unit 11 generates an explanatory text vector from the extracted text information, and generates a multi-dimensional content value vector from the explanatory text vector. The dimensions of the content value vector are, for example, the following 10. Dimension 1: Fame (This indicates that the user wants to go to a famous place) Dimension 2: Challenge (This indicates that the user wants to do something challenging) Dimension 3: Sympathy (This indicates that the user wants to give benefits to others) Dimension 4: Low price (This indicates that the user wants to go to a place with a low price) Dimension 5: Evaluation (This indicates that the user wants to go to a place with a high evaluation by others) Dimension 6: Health (This indicates that the user wants to be healthy) Dimension 7: Preference (This indicates that the user wants to do what they like) Dimension 8: Seasonality (This indicates that the user can only do it now) Dimension 9: Attachment (This indicates that the user wants to go to a familiar place) Dimension 10: Connection (This indicates that the user wants to connect with someone)

[0027] In generating the content value vector, the following algorithm may be used. · Data augmentation · Sentence-BERT · SBERT-WK · bert-fine · Fine-tuning

[0028] The algorithm used in the generation is not limited to those described above and may be arbitrarily selected from conventionally well-known ones. The algorithm may utilize AI (Artificial Intelligence). A detailed explanation of the processing executed by the algorithm is omitted here.

[0029] For example, the control unit 11 automatically extracts representative words from text information and creates a word set (vocabulary set) in generating the content value vector. The control unit 11 uses morphological analysis to fix the part of speech to be extracted, such as nouns or objects. The control unit 11 extracts words so as to be as dispersed as possible within the vector space output in generating the explanatory text vector. The representative words are used for generating learning data.

[0030] The control unit 11 performs the following scoring for each of the extracted words (here, "fashionable", "quiet", "satisfying") manually by the user. Fashionable Extroversion 80% Quiet Introversion 60% Satisfying Emphasis on cost performance 90%

[0031] Furthermore, the control unit 11 generates the following explanatory text vectors using the combination logic of data augmentation. It is a quiet and satisfying store (0.0, 0.6, 0.9) The store was fashionable (0.8, 0.0, 0.0)

[0032] The control unit 11 generates a content value vector in multiple dimensions from the explanatory text vector. The control unit 11 can learn a model for generating a content value vector from the content of the store.

[0033] The control unit 11 calculates the user's personal parameters from the user's input information. The input information includes, for example, information input in a questionnaire. Specifically, the input information includes at least one of discrete data, numerical data, and numerical and discrete data. For example, when there is a question in a questionnaire such as "How many stars would you give?", the answer to that question is numerical data. The answer to a question such as "Select all that apply" is discrete data because it is a vector consisting of 1 or 0. The answer to a question such as "Please tell me your highest level of education" is numerical and discrete data because it is a combination of a numerical value and a one-hot vector.

[0034] The control unit 11 aggregates the input information horizontally (i.e., for the same respondent) and vertically (i.e., by grouping multiple respondents). In the vertical aggregation, processes such as averaging, dispersion, and normalization may be performed. The personal parameters indicate, for example, a health orientation score, an extroversion score, a communication orientation score, a health orientation score, etc.

[0035] The control unit 11 generates a user preference value vector in the same multiple dimensions as those used in the content value vector from the calculated personal parameters. In the generation of the user preference value vector, any estimation model using an algorithm such as Feature Engineering may be used. The control unit 11 can determine which dimension values of the user preference value vector are relatively high for each user. The control unit 11 can learn a model for generating a user preference value vector from the personal parameters.

[0036] As shown in FIG. 4, the control unit 11 stores, in association with the ID, the content value vector of each content and the user preference value vector of each user in the similarity DB. The control unit 11 numerically determines the similarity (affinity) between the content value vector and the user preference value vector. The similarity may be determined by a cosine similarity, Euclidean distance, mean squared error, or a machine learning model based on a ranking algorithm.

[0037] When the user terminal 2 receives a request operation from the user at the display unit 24 or the input unit 25, it transmits the request to the information processing apparatus 1. The request may be, for example, "want to relax", "want to eat", or "want to do sports". When the control unit 11 of the information processing apparatus 1 receives the request, it extracts store options. The control unit 11 generates a content value vector for each option.

[0038] The control unit 11 generates a user preference value vector for the user of the user terminal 2. The control unit 11 presents one or more stores corresponding to the content value vector having the highest similarity (or higher than the reference value) with the calculated user preference value vector to the user terminal 2. The similarity here corresponds to the level of happiness.

[0039] When the control unit 21 of the user terminal 2 receives a selection from the user for one of the options, it proposes to make a reservation for the store corresponding to the selected option. The control unit 21 can receive a reservation request from the user and make a reservation for the store.

[0040] As an additional example or an alternative example, when the control unit 21 receives a request operation from a plurality of users (for example, group behavior members) belonging to the same group, it may determine the similarity for each user and notify the determination result to the user terminals of each user.

[0041] Referring to FIG. 5, an information processing method by the information processing apparatus 1 will be described.

[0042] In step S1, the control unit 11 extracts text information related to the store from the store's content. In step S2, the control unit 11 generates a content value vector in multiple dimensions from the text information.

[0043] In step S3, the control unit 11 calculates the user's personal parameters from the user's input information. In step S4, the control unit 11 generates a user preference value vector in multiple dimensions from the personal parameters. Steps S1 and S2, and steps S3 and S4 may be executed in the reverse order.

[0044] In step S5, the control unit 11 determines the similarity between the content value vector and the user preference value vector.

[0045] As described above, according to this embodiment, the control unit 11 extracts text information related to the store from the store's content, generates a content value vector in multiple dimensions from the text information, calculates the user's personal parameters from the user's input information, generates a user preference value vector in multiple dimensions from the personal parameters, and determines the similarity between the content value vector and the user preference value vector. With this configuration, the information processing apparatus 1 can evaluate the store in multiple dimensions, so that it can evaluate content that more closely matches the user's preferences.

[0046] Also according to this embodiment, the similarity is determined by cosine similarity, Euclidean distance, mean squared error, or a machine learning model using a ranking algorithm. With this configuration, the information processing apparatus 1 can determine the similarity more accurately.

[0047] Also according to this embodiment, the operation of the control unit 11 includes acquiring the store's content and generating a content value vector using a predetermined algorithm. With this configuration, the information processing apparatus 1 can quantify the content value of the store.

[0048] Also according to the present embodiment, the operation of the control unit 11 includes learning a model that generates a user preference value vector from the person parameters. With this configuration, the information processing apparatus 1 can improve the generation accuracy of the user preference value vector.

[0049] Also according to the present embodiment, the operation of the control unit 11 includes, when receiving a request operation from the user, notifying the user of one or more options of the store from the determined high similarity. With this configuration, the information processing apparatus 1 can evaluate the user happiness for each option proposed to the user. The information processing apparatus 1 can further evaluate individual elements of the user happiness by using a multi-dimensional vector.

[0050] It should be noted that although the present disclosure has been described based on the drawings and examples, those skilled in the art may make various modifications and alterations based on the present disclosure. In addition, changes can be made without departing from the spirit of the present disclosure. For example, the functions included in each means or each step can be rearranged so as not to be logically contradictory, and a plurality of means or steps can be combined into one or divided.

[0051] For example, in the above embodiment, a program for executing all or part of the functions or processes of the information processing apparatus 1 can be recorded on a computer-readable recording medium. The computer-readable recording medium includes a non-transitory computer-readable medium, and is, for example, a magnetic recording device, an optical disk, a magneto-optical recording medium, or a semiconductor memory. The distribution of the program can be performed, for example, by selling, transferring, or lending a portable recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory) on which the program is recorded. The distribution of the program may also be performed by storing the program in the storage of an arbitrary server and transmitting the program from the arbitrary server to another computer. The program may also be provided as a program product. The present disclosure can also be realized as a program executable by a processor.

Explanation of Signs

[0052] 1 Information processing apparatus

Claims

1. An information processing apparatus including a control unit, wherein the control unit extracts text information regarding the store from the content of the store, extracts words from the text information and manually scores the words by a user who performs the scoring, generates a content value vector in multiple dimensions using the word and the result of the scoring for the word, calculates a personal parameter of the user who makes the input from the input information of the user who makes the input, generates a user preference value vector in the multiple dimensions from the personal parameter, determines the similarity between the content value vector and the user preference value vector, and executes an operation including the above, the information processing apparatus.

2. In the information processing apparatus according to Claim 1, the similarity is determined by a cosine similarity, a Euclidean distance, a mean squared error, or a machine learning model using a ranking algorithm, the information processing apparatus.

3. In the information processing apparatus according to Claim 1, the operation includes generating the content value vector using a predetermined algorithm from the word and the result of the scoring for the word, the information processing apparatus.

4. In the information processing apparatus according to Claim 1, the operation includes learning a model for generating the user preference value vector from the personal parameter, the information processing apparatus.

5. In the information processing apparatus according to Claim 1, the operation includes, when receiving a request operation from an arbitrary user, notifying the arbitrary user of one or more options of the store based on the determined high similarity, the information processing apparatus.

Images