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

Abstract

To provide an information processing device capable of integrating a model for a similar product search and a model for a wrong product search in a commodity search field of an online shopping mall.SOLUTION: An information processing device comprises: an acquisition unit which acquires learning data for a similar product search and learning data for a wrong product search; and an integrated learning unit 133 which conducts multitask learning of a task for the similar product search and a task for the wrong product search using the learning data for the similar product search and the learning data for the wrong product search.SELECTED DRAWING: Figure 6

Description

Technical Field

[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 providing a multi-task learning network and a method based on the multi-task learning network have been disclosed (for example, Patent Document 1). For example, in this technique, a task network includes a plurality of sub-task networks, the plurality of sub-task networks generate corresponding feature vectors for the same input, a merging network performs concatenation on the corresponding feature vectors, and optimization processing is performed on the concatenated feature vectors to obtain a final output.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, there is room for improvement in the above conventional technology. For example, in the above conventional technology, although a plurality of sub-task networks perform concatenation on feature vectors and optimization processing is performed on the concatenated feature vectors to obtain a final output, there is room for improvement in integrating models that perform different tasks. For example, it is desired to integrate a model for performing similar product search and a model for performing error search in the field of product search in an online shopping mall.

[0005] The present application has been made in view of the above, and an object thereof is to integrate a model for performing similar product search and a model for performing error search in the field of product search in an online shopping mall. [Means for solving the problem]

[0006] The information processing device according to the present application comprises: an acquisition unit that acquires learning data for similar product search and learning data for product difference detection; and an integrated learning unit that uses the learning data for similar product search and the learning data for product difference detection to perform multitask learning for a similar product search task and a product difference detection task. The integrated learning unit, in the learning for similar product search, inputs the text of the product master and product data into the BERT model and performs a task of multi-class classification with products as the class for similar product search. In the learning for product difference detection, it inputs the text of the product data into the BERT model and performs a task of binary classification of identical products for product difference detection. It is characterized by the following: [Effects of the Invention]

[0007] According to one embodiment, a model for similar product search and a model for finding differences can be integrated in the field of product search for online shopping malls. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 shows an example of the configuration of an information processing system according to an embodiment. [Figure 2] Figure 2 is an explanatory diagram illustrating the current learning methods. [Figure 3] Figure 3 is an explanatory diagram illustrating the overview of model integration for similar product search and spot-the-difference using multitasking learning. [Figure 4] Figure 4 is an explanatory diagram illustrating the overview of how the underlying BERT dataset is shared. [Figure 5] Figure 5 shows an example of the configuration of a terminal device according to this embodiment. [Figure 6] Figure 6 shows an example of the configuration of a server device according to the embodiment. [Figure 7] Figure 7 is a flowchart showing the processing procedure according to the embodiment. [Figure 8] Figure 8 shows an example of a hardware configuration. [Modes for carrying out the invention]

[0009] The following describes in detail, with reference to the drawings, embodiments for implementing the information processing device, information processing method, and information processing program according to the present application (hereinafter referred to as "embodiments"). Note that these embodiments do not limit the information processing device, information processing method, and information processing program according to the present application. Furthermore, the same parts are denoted by the same reference numerals in the following embodiments, and redundant descriptions are omitted.

[0010] [1. Example of an information processing system configuration] First, the configuration of the information processing system 1, which includes the server device 100 according to the embodiment, will be described using Figure 1. Figure 1 is a diagram showing an example of the configuration of the information processing system 1 according to the embodiment. As shown in Figure 1, the information processing system 1 according to the embodiment includes a terminal device 10 and a server device 100. These various devices are connected to each other via a network N, either by wire or wireless means, enabling communication. As a result, the terminal device 10 can cooperate with the server device 100. The network N is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network) such as the Internet.

[0011] Terminal device 10 is an information processing device used by user U. For example, terminal device 10 may be a smart device such as a smartphone or tablet, a mobile phone such as a feature phone, a PC (Personal Computer), a PDA (Personal Digital Assistant), a game console or AV equipment with communication functions, an information appliance or digital appliance, a car navigation system, a wearable device such as a smartwatch, head-mounted display, or smart glasses. Alternatively, terminal device 10 may be a house or building compatible with the Internet of Things (IoT), a car, a home appliance, or an electronic device.

[0012] In this embodiment, the terminal device 10 is a smart device such as a smartphone or tablet used by user U, and is a mobile terminal device capable of communicating with any server device via a wireless communication network such as LTE (Long Term Evolution), 4G (4th Generation), or 5G (5th Generation). The terminal device 10 also has a screen such as a liquid crystal display with touch panel functionality, and accepts various operations on displayed data such as content from user U using a finger or stylus, such as tapping, sliding, and scrolling. Operations performed on the area of ​​the screen where content is displayed may also be considered as operations on the content. Furthermore, the terminal device 10 may be an information processing device such as a desktop PC (Personal Computer) or a notebook PC, not just a smart device.

[0013] Furthermore, the terminal device 10 can connect to the network N via wireless communication networks such as LTE, 4G, and 5G, or via short-range wireless communication such as Bluetooth® and Wi-Fi (Local Area Network), and communicate with the server device 100.

[0014] The server device 100 is, for example, a computer such as a PC or blade server, or a mainframe or workstation. The server device 100 may also be implemented through cloud computing.

[0015] In this embodiment, the server device 100 is an information processing device that works in conjunction with each user U's terminal device 10 and provides each user U's terminal device 10 with API (Application Programming Interface) services for various applications (hereinafter referred to as "apps") and various data, and is implemented by a computer or cloud system.

[0016] Further, the server device 100 may be an information processing device that provides some kind of web service online to each terminal device 10 of each user U. For example, as a web service, the server device 100 may provide services such as Internet connection, search service, SNS (Social Networking Service), e-commerce (EC: Electronic Commerce), electronic payment, online game, online banking, online trading, accommodation and ticket reservation, video and music distribution, news, map, route search, route guidance, route information, operation information, weather forecast, etc. Actually, the server device 100 may cooperate with various servers that provide the above web services and mediate the web services, or may be in charge of the processing of the web services.

[0017] In addition, the server device 100 can acquire user information regarding the user U. For example, the server device 100 acquires information regarding the attributes of the user U, such as the gender, age, and residential area of the user U. Then, the server device 100 stores and manages the information regarding the attributes of the user U together with the identification information (such as user ID) indicating the user U.

[0018] In addition, the server device 100 acquires various types of history information (log data) indicating the actions of the user U from the terminal device 10 of the user U or from various servers or the like based on the user ID or the like. For example, the server device 100 acquires a location history, which is a history of the location and time of the user U, from the terminal device 10. In addition, the server device 100 acquires a search history, which is a history of search queries input by the user U, from a search server (search engine). In addition, the server device 100 acquires a browsing history, which is a history of content browsed by the user U, from a content server. In addition, the server device 100 acquires a purchase history (settlement history), which is a history of product purchases and settlement processes of the user U, from an e-commerce server or a settlement processing server. In addition, the server device 100 may acquire a listing history or a sales history, which is a history of the user U's listings on the marketplace, from an e-commerce server or a settlement processing server. In addition, the server device 100 acquires a posting history, which is a history of posts of the user U, from a posting server or an SNS server that provides a word-of-mouth posting service. Note that each of the above various servers or the like may be the server device 100 itself. That is, the server device 100 may function as each of the above various servers or the like.

[0019] In addition, the number of each device included in the information processing system 1 shown in FIG. 1 is not limited to that shown. For example, in FIG. 1, only one terminal device 10 is shown for simplification of illustration, but this is merely an example and is not limited, and two or more may be provided.

[0020] [2. Multi-task learning method of BERT] In the present embodiment, a multi-task learning method of BERT (Bidirectional Encoder Representations from Transformers) implemented by the server device 100 will be described. Currently, in research and development for the e-commerce field of product search in an online shopping mall, there are two products: "similar product search" and "finding mistakes". BERT is used in the two products.

[0021] (1) Similar product search Vector searches are performed using NGT (Vector Neighborhood Search software) for nearby neighbor searches in high-dimensional vector data. BERT is used for embedding text (the process of converting words, sentences, etc., into vector representations) to construct the index. For example, BERT is used for embedding product names. NGT is used when performing vector searches using the embedded vectors. Prior to performing vector searches, product data from online shopping malls is embedded to create an index. In this process, a dense vector representing a specific word is associated with that word's embedding.

[0022] (2) Find the difference Given a pair of products, we need to determine if that pair consists of identical products. BERT is used to determine if products are identical based on text, and there are two methods for doing so.

[0023] (A) cos similarity The cosine similarity (cosine similarity) is calculated for both the product name and the brand name, which are embedded separately, and the determination is made based on the similarity.

[0024] (B) Binary Classification The system takes a set of product name and product name as input and performs binary classification to determine if they are the same product.

[0025] Currently, the cosine similarity method (A) above yields higher accuracy, but I often receive comments that it's more intuitive for the binary classification method (B) above to yield higher accuracy.

[0026] If we decide to use binary classification for identifying identical products in a spot-the-difference puzzle, we would end up with two BERT models: one for embedding and one for binary classification. From a research and development human resource perspective, there is a desire to integrate the BERT model for embedding and the BERT model for binary classification.

[0027] To clarify the learning methods, we will now describe the current learning methods and a proposed method for integrating models using multi-task learning.

[0028] [2-1. Current learning methods] Figure 2 is an explanatory diagram illustrating the overview of the current learning method. In the current learning method, learning for embedding and learning for identical product detection are performed separately.

[0029] For example, as shown in Figure 2, in training for Embedding for similar product search, the text of the product master and product data is input into the BERT model, and the model is trained to perform Task A, which involves multi-class classification with products as the class. In this case, the training data for Embedding uses the product master and product data associated with the products.

[0030] Furthermore, for training in the same-product detection task for spot-the-difference games, the text of the product data is input into the BERT model, and the model is trained on Task B, which involves binary classification of identical products. In this case, the product data used as training data for same-product detection includes both product data that is associated with a specific product and product data that is not associated with a specific product.

[0031] In the current learning method, the training for embedding and the training for identical product detection are independent of each other. Since the training data for embedding and the training for identical product detection are different, the parameters of each BERT model will be different after training.

[0032] [2-2. Model Integration through Multitask Learning] Figure 3 is an explanatory diagram illustrating the overview of model integration for similar product search and product difference detection using multitask learning. In this embodiment, multitask learning is used to integrate BERT models for similar product search and product difference detection, and to improve their accuracy.

[0033] For example, as shown in Figure 3, in the integration of models for similar product search and spot the difference using multitask learning, the text of the product master and product data used for embedding in similar product search and the text of the product data used for identical product determination in spot the difference are input into the same BERT model, and multitask learning is performed for multi-class classification with products as classes (Task A) and binary classification of identical products (Task B).

[0034] For example, if the text from the product master is pre-entered into the model, and then the text from the product data is entered into the model, the model will simultaneously perform multi-class classification with products as the class (Task A) and binary classification of identical products (Task B).

[0035] Alternatively, if the product master and product data text used for embedding for similar product search are input into the BERT model, a multi-class classification (Task A) with products as the class is performed. If the product data text used for training for identical product detection in a spot-the-difference application is input into the BERT model, binary classification of identical products (Task B) is performed.

[0036] Furthermore, to improve the accuracy of embedding and identifying identical products, the BERT model may be trained to pay attention to factors such as capacity, quantity, and brand. For example, by adding a task to classify the capacity and quantity of products in multi-task learning, the model can be trained to predict the capacity and quantity of products. Similarly, by adding a task to classify the brand of products in multi-task learning, the model can be trained to predict the brand of products. Additionally, if there is other data that can be used for training, tasks can be added.

[0037] [2-3. How to share the base BERT dataset] Figure 4 is an explanatory diagram illustrating the overview of the method for sharing the base BERT model. While this method of sharing the base BERT model resembles model integration other than multi-task learning, it is not strictly model integration. It simply uses a model trained for similar product search as the base model for the error detection task.

[0038] For example, as shown in Figure 4, in training for Embedding for similar product search, the text of the product master and product data is input into the BERT model, and multi-class classification (Task A) is performed with products as the class.

[0039] Furthermore, in model integration other than multitask learning, the parameters obtained from training for embedding for similar product search are reused for training for identical product detection in difference-finding. In other words, the parameters of the BERT model trained for embedding for similar product search are applied to the BERT model trained for identical product detection in difference-finding.

[0040] Then, for learning to identify identical products for the "find the difference" task, the text of the product data is input into a BERT model that reuses the parameters obtained from the learning for embedding for similar product search, and binary classification of identical products (Task B) is performed.

[0041] Alternatively, after creating a BERT model trained for similar product search, a binary classification layer can be added to the BERT model, and the parameters of the BERT model can be fixed for additional training to detect product differences.

[0042] [2-4. Characteristics] As described above, the BERT multitask learning method according to this embodiment is characterized by integrating a model that performs similar product search and a model that performs error detection in the field of product search for online shopping malls. In other words, it integrates a model for determining similar products with a model for determining whether the association between products and the database is correct.

[0043] The server device 100 learns a model that estimates which product in the database corresponds to a listed product, based on the product information entered by the seller. The server device 100 also uses the learned model to determine whether a product and a product are identical when product information and product information are entered.

[0044] For example, the server device 100 learns the task of performing a vector search and the task of checking whether the association is incorrect through multitasking learning.

[0045] Furthermore, additional tasks will be added, such as learning to guess the product's capacity and quantity, and learning to identify the brand. Additionally, it would be beneficial to have the students perform subtasks that are considered useful for product identification.

[0046] Furthermore, the server device 100 extracts the problem to be considered in Task A as Task B and repeatedly trains the model. The task may be binary classification or multi-class classification.

[0047] In this embodiment, training is performed gradually in mini-batches, but it is also possible to do it roughly and comprehensively. Alternatively, separate BERT datasets for products and goods can be set up and combined using the CONCAT function (CONCATENATE function) to concatenate strings.

[0048] [3. Example of terminal device configuration] Next, the configuration of the terminal device 10 will be explained using Figure 5. Figure 5 is a diagram showing an example of the configuration of the terminal device 10. As shown in Figure 5, the terminal device 10 comprises a communication unit 11, a display unit 12, an input unit 13, a positioning unit 14, a sensor unit 20, a control unit 30 (controller), and a storage unit 40.

[0049] (Communications Section 11) The communication unit 11 is connected to the network N by wire or wireless connection and transmits and receives information to and from the server device 100 via the network N. For example, the communication unit 11 can be implemented using a NIC (Network Interface Card) or an antenna.

[0050] (Display section 12) The display unit 12 is a display device that displays various information such as location information. For example, the display unit 12 may be a liquid crystal display (LCD) or an organic electro-luminescent display (OLED). The display unit 12 may also be a touch panel display, but is not limited to this.

[0051] (Input section 13) The input unit 13 is an input device that receives various operations from the user U. For example, the input unit 13 has buttons for inputting characters, numbers, etc. The input unit 13 may also be an input / output port (I / O port) or a USB (Universal Serial Bus) port. If the display unit 12 is a touch panel display, a part of the display unit 12 functions as the input unit 13. The input unit 13 may also be a microphone that receives voice input from the user U. The microphone may be wireless.

[0052] (Positioning unit 14) The positioning unit 14 receives signals (radio waves) transmitted from GPS (Global Positioning System) satellites and, based on the received signals, acquires position information (e.g., latitude and longitude) indicating the current position of the terminal device 10. In other words, the positioning unit 14 determines the position of the terminal device 10. Note that GPS is just one example of a GNSS (Global Navigation Satellite System).

[0053] Furthermore, the positioning unit 14 can determine its position using various methods other than GPS. For example, the positioning unit 14 may use various communication functions of the terminal device 10 to determine its position as an auxiliary positioning means for position correction, etc., as described below.

[0054] (Wi-Fi positioning) For example, the positioning unit 14 determines the location of the terminal device 10 by utilizing the Wi-Fi® communication function of the terminal device 10 and the communication network provided by each telecommunications company. Specifically, the positioning unit 14 determines the location of the terminal device 10 by performing Wi-Fi communication, etc., and determining the distance to nearby base stations and access points.

[0055] (Beacon positioning) Furthermore, the positioning unit 14 may determine the location using the Bluetooth® function of the terminal device 10. For example, the positioning unit 14 determines the location of the terminal device 10 by connecting to a beacon transmitter connected via the Bluetooth® function.

[0056] (Geomagnetic positioning) Furthermore, the positioning unit 14 determines the position of the terminal device 10 based on the geomagnetic pattern of the structure, which has been measured in advance, and the geomagnetic sensor provided by the terminal device 10.

[0057] (RFID positioning) Furthermore, if, for example, the terminal device 10 is equipped with an RFID (Radio Frequency Identification) tag function equivalent to that of a contactless IC card used at a train station ticket gate or in a store, or if it is equipped with a function to read RFID tags, the location where it was used will be recorded along with the information on the payment or other transactions made by the terminal device 10. The positioning unit 14 may determine the location of the terminal device 10 by acquiring such information. Alternatively, the location may be determined by an optical sensor or infrared sensor equipped in the terminal device 10.

[0058] The positioning unit 14 may, if necessary, determine the position of the terminal device 10 using one or a combination of the positioning means described above.

[0059] (Sensor unit 20) The sensor unit 20 includes various sensors mounted on or connected to the terminal device 10. The connection can be wired or wireless. For example, the sensors may be detection devices other than the terminal device 10, such as wearable devices or wireless devices. In the example shown in Figure 5, the sensor unit 20 includes an acceleration sensor 21, a gyro sensor 22, a barometric pressure sensor 23, a temperature sensor 24, a sound sensor 25, a light sensor 26, a magnetic sensor 27, and an image sensor (camera) 28.

[0060] The sensors 21-28 described above are merely examples and not limiting. In other words, the sensor unit 20 may be configured to include some of the sensors 21-28, or it may include other sensors such as humidity sensors in addition to or instead of the sensors 21-28.

[0061] The acceleration sensor 21 is, for example, a 3-axis acceleration sensor and detects the physical movement of the terminal device 10, such as its direction of movement, velocity, and acceleration. The gyro sensor 22 detects the physical movement of the terminal device 10, such as its tilt in the three axes, based on its angular velocity. The barometric pressure sensor 23 detects the atmospheric pressure around the terminal device 10, for example.

[0062] Since the terminal device 10 is equipped with the acceleration sensor 21, gyroscope 22, barometric pressure sensor 23, etc., it becomes possible to determine the position of the terminal device 10 using technologies such as pedestrian dead-reckoning (PDR) that utilize these sensors 21 to 23. This makes it possible to obtain indoor location information that is difficult to obtain with positioning systems such as GPS.

[0063] For example, a pedometer using an accelerometer 21 can calculate the number of steps, walking speed, and distance walked. Additionally, a gyroscope 22 can be used to determine the user U's direction of movement, gaze direction, and body tilt. Furthermore, the barometric pressure detected by the barometric pressure sensor 23 can be used to determine the altitude and floor number of the user U's terminal device 10.

[0064] The temperature sensor 24 detects, for example, the ambient temperature around the terminal device 10. The sound sensor 25 detects, for example, the ambient sound around the terminal device 10. The light sensor 26 detects the ambient illumination around the terminal device 10. The magnetic sensor 27 detects, for example, the Earth's magnetic field around the terminal device 10. The image sensor 28 captures an image of the area around the terminal device 10.

[0065] The aforementioned pressure sensor 23, temperature sensor 24, sound sensor 25, light sensor 26, and image sensor 28 can detect the surrounding environment and conditions of the terminal device 10 by detecting atmospheric pressure, temperature, sound, and illuminance, respectively, and by capturing images of the surroundings. Furthermore, it becomes possible to improve the accuracy of the location information of the terminal device 10 based on the surrounding environment and conditions.

[0066] (Control Unit 30) The control unit 30 includes, for example, a microcomputer having a CPU (Central Processing Unit), ROM (Read Only Memory), RAM, input / output ports, and various circuits. Alternatively, the control unit 30 may be composed of hardware such as an integrated circuit (ASIC) or FPGA (Field Programmable Gate Array). The control unit 30 includes a transmission unit 31, a reception unit 32, and a processing unit 33.

[0067] (Transmitter 31) The transmission unit 31 can transmit various information, such as information input by the user U using the input unit 13, various information detected by sensors 21-28 mounted on or connected to the terminal device 10, and location information of the terminal device 10 determined by the positioning unit 14, to the server device 100 via the communication unit 11.

[0068] (Receiving unit 32) The receiving unit 32 can receive various information provided by the server device 100, as well as requests for various information from the server device 100, via the communication unit 11.

[0069] (Processing 33) The processing unit 33 controls the entire terminal device 10, including the display unit 12. For example, the processing unit 33 can output and display various information transmitted by the transmission unit 31 and various information received from the server device 100 by the reception unit 32 to the display unit 12.

[0070] (Storage unit 40) The storage unit 40 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or by storage devices such as HDD (Hard Disk Drive), SSD (Solid State Drive), and optical discs. Various programs and various data are stored in this storage unit 40.

[0071] [4. Example of Server Device Configuration] Next, the configuration of the server device 100 according to the embodiment will be described using Figure 6. Figure 6 is a diagram showing an example of the configuration of the server device 100 according to the embodiment. As shown in Figure 6, the server device 100 includes a communication unit 110, a storage unit 120, and a control unit 130.

[0072] (Communications Department 110) The communication unit 110 is implemented, for example, by a NIC (Network Interface Card). The communication unit 110 is connected to the network N by wire or wireless connection.

[0073] (Storage unit 120) The storage unit 120 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or by storage devices such as HDDs, SSDs, and optical discs. The storage unit 120 may store identification information (such as a user ID) indicating user U, as well as attribute information and history information (log data) of user U.

[0074] The memory unit 120 may also store product master data and product data. Furthermore, the memory unit 120 may store training datasets. Additionally, the memory unit 120 may store a trained BERT model.

[0075] (Control unit 130) The control unit 130 is a controller, and is realized by executing various programs (corresponding to an example of an information processing program) stored in the internal memory of the server device 100 using a memory area such as RAM as a working area, for example, by a CPU (Central Processing Unit), MPU (Micro Processing Unit), ASIC (Application Specific Integrated Circuit), or FPGA (Field Programmable Gate Array). In the example shown in Figure 6, the control unit 130 has an acquisition unit 131, a creation unit 132, an integrated learning unit 133, an inference unit 134, and a provision unit 135.

[0076] (Acquisition part 131) The acquisition unit 131 acquires the search query entered by the user U. For example, when the user U enters a search query into a search engine or the like and performs a keyword search, the acquisition unit 131 acquires the search query via the communication unit 110. In other words, the acquisition unit 131 acquires the keyword entered by the user U into the search box of a search engine, website, or application via the communication unit 110.

[0077] Furthermore, the acquisition unit 131 acquires user information about user U via the communication unit 110. For example, the acquisition unit 131 acquires identification information (such as user ID), location information, and attribute information of user U from user U's terminal device 10. The acquisition unit 131 may also acquire identification information and attribute information of user U when user U is registered. The acquisition unit 131 then stores the user information in the storage unit 120.

[0078] Furthermore, the acquisition unit 131 acquires various historical information (log data) indicating the user U's actions via the communication unit 110. For example, the acquisition unit 131 acquires various historical information indicating the user U's actions from the user U's terminal device 10, or from various servers based on the user ID, etc. The acquisition unit 131 then stores the various historical information in the storage unit 120.

[0079] Furthermore, the acquisition unit 131 acquires learning data for similar product search and learning data for product difference detection.

[0080] (Creation section 132) The creation unit 132 may create a training dataset based on the source data. For example, the creation unit 132 may create training data for similar product search from product master and product data. Alternatively, the creation unit 132 may create training data for product difference detection from product data. Note that if the acquisition unit 131 has acquired the training data for similar product search and the training data for product difference detection, the creation unit 132 does not need to create the training data for similar product search and the training data for product difference detection. The creation unit 132 may also be integrated with the acquisition unit 131. For example, creating the training data for similar product search and the training data for product difference detection may be considered as having acquired the training data for similar product search and the training data for product difference detection.

[0081] (Integrated Learning Section 133) The integrated learning unit 133 uses training data for similar product search and training data for product difference detection to perform multitask learning for similar product search and product difference detection tasks.

[0082] For example, the integrated learning unit 133 inputs training data for similar product search and training data for product difference detection into the same model, and performs training for the similar product search task and the product difference detection task in parallel.

[0083] Alternatively, the integrated learning unit 133 learns the task of similar product search when learning data for similar product search is input to the model, and learns the task of product difference detection when learning data for product difference detection is input to the same model.

[0084] Alternatively, the integrated learning unit 133 learns the task of similar product search when training data for similar product search is input into the model, and then reuses the model's parameters to learn the task of product difference detection when training data for product difference detection is input into the model to which the parameters have been applied.

[0085] Alternatively, the integrated learning unit 133 learns the task of similar product search when it receives training data for similar product search into the model, adds a task for product difference detection to the model, fixes the model parameters, and performs additional learning for product difference detection.

[0086] Furthermore, in the learning process for similar product search, the integrated learning unit 133 inputs the text of the product master and product data into the BERT model and performs a task of multi-class classification with products as the class.

[0087] Furthermore, in the training for product error detection, the integrated learning unit 133 inputs the text of the product data into the BERT model and performs the task of binary classification of identical products.

[0088] Furthermore, the integrated learning unit 133 performs multitasking learning by adding a task to classify the capacity and number of products. Furthermore, the integrated learning unit 133 performs multitasking learning by adding a task to classify the brand of the products.

[0089] (Inference part 134) The inference unit 134 performs the task of similar product search using input data for similar product search and a trained model. For example, the inference unit 134 inputs the text of the product data into the BERT model and performs multi-class classification with products as the class.

[0090] Furthermore, the inference unit 134 performs a product difference detection task using input data and a trained model. For example, the inference unit 134 inputs the text of the product data into a BERT model and performs binary classification of identical products.

[0091] Furthermore, the inference unit 134 inputs the text of the product data into the BERT model and performs the task of predicting the capacity and number of items in a product. In addition, the inference unit 134 inputs the text of the product data into the BERT model and performs the task of predicting the brand of a product.

[0092] (Provider 135) The provisioning unit 135 provides the inference results from the inference unit 134 to the user U's terminal device 10 or an external server device via the communication unit 110. For example, the provisioning unit 135 provides the results of multi-class classification with products as classes, or the results of binary classification of the same product.

[0093] Alternatively, the provisioning unit 135 provides search results based on the inference results from the inference unit 134 to the user U's terminal device 10 via the communication unit 110, in response to a search query from the user U's terminal device 10.

[0094] [5. Processing Procedure] Next, the processing procedure by the server device 100 according to the embodiment will be described using Figure 7. Figure 7 is a flowchart of the processing procedure according to the embodiment. Note that the processing procedure shown below is repeatedly executed by the control unit 130 of the server device 100.

[0095] For example, as shown in Figure 7, the acquisition unit 131 of the server device 100 acquires product master and product data (step S101).

[0096] Next, the creation unit 132 of the server device 100 creates training data for similar product search from the product master and product data text, and creates training data for product difference detection from the product data text (step S102).

[0097] Next, the training data for similar product search and the training data for product difference detection from the server device 100 are input into the BERT model to perform multitasking learning for the similar product search task and the product difference detection task (step S103).

[0098] Next, the inference unit 134 of the server device 100 inputs input data for similar product search into the trained model and performs a similar product search task, and inputs input data for product difference detection and performs a product difference detection task (step S104). In practice, the inference unit 134 may perform the similar product search task and the product difference detection task in parallel using the same input data.

[0099] Next, the provision unit 135 of the server device 100 provides the results of each task (inference results) performed by the inference unit 134 to the user U's terminal device 10 or an external server device via the communication unit 110 (step S105).

[0100] [6. Variant Example] The terminal device 10 and server device 100 described above may be implemented in various other forms besides those of the embodiment described above. Therefore, the following describes modifications of the embodiment.

[0101] In the above embodiment, some or all of the processing performed by the server device 100 may actually be performed by the terminal device 10. For example, the processing may be completed in a standalone manner (by the terminal device 10 alone). In this case, the terminal device 10 is assumed to have the functions of the server device 100 in the above embodiment. Furthermore, in the above embodiment, since the terminal device 10 is in cooperation with the server device 100, from the perspective of the user U, it appears as if the processing of the server device 100 is also being performed by the terminal device 10. In other words, from another perspective, it can be said that the terminal device 10 is equipped with the server device 100.

[0102] [7. Effects] As described above, the information processing device (server device 100) according to the present application is characterized by comprising an acquisition unit 131 that acquires learning data for similar product search and learning data for product difference detection, and an integrated learning unit 133 that performs multitask learning of a similar product search task and a product difference detection task using the learning data for similar product search and the learning data for product difference detection.

[0103] For example, the integrated learning unit 133 inputs training data for similar product search and training data for product difference detection into the same model, and performs training for the similar product search task and the product difference detection task in parallel.

[0104] Alternatively, the integrated learning unit 133 learns the task of similar product search when learning data for similar product search is input to the model, and learns the task of product difference detection when learning data for product difference detection is input to the same model.

[0105] Alternatively, the integrated learning unit 133 learns the task of similar product search when training data for similar product search is input into the model, and then reuses the model's parameters to learn the task of product difference detection when training data for product difference detection is input into the model to which the parameters have been applied.

[0106] Alternatively, the integrated learning unit 133 learns the task of similar product search when it receives training data for similar product search into the model, adds a task for product difference detection to the model, fixes the model parameters, and performs additional learning for product difference detection.

[0107] Furthermore, in the learning process for similar product search, the integrated learning unit 133 inputs the text of the product master and product data into the BERT model and performs a task of multi-class classification with products as the class.

[0108] Furthermore, in the training for product error detection, the integrated learning unit 133 inputs the text of the product data into the BERT model and performs the task of binary classification of identical products.

[0109] Furthermore, the integrated learning unit 133 performs multitasking learning by adding a task to classify the capacity and number of products.

[0110] Furthermore, the integrated learning unit 133 performs multitasking learning by adding the task of classifying product brands.

[0111] By any or a combination of the above-described processes, the information processing device according to the present invention can integrate a model for similar product search and a model for error detection in the field of product search for online shopping malls.

[0112] [8. Hardware Configuration] Furthermore, the terminal device 10 and server device 100 according to the above-described embodiment are realized by a computer 1000 having a configuration such as that shown in Figure 8. The following explanation will use the server device 100 as an example. Figure 8 is a diagram showing an example of the hardware configuration. The computer 1000 is connected to an output device 1010 and an input device 1020, and has a configuration in which an arithmetic unit 1030, a primary storage device 1040, a secondary storage device 1050, an output interface 1060, an input interface 1070, and a network interface 1080 are connected by a bus 1090.

[0113] The arithmetic unit 1030 operates based on programs stored in the primary storage device 1040 and the secondary storage device 1050, as well as programs read from the input device 1020, and executes various processes. The arithmetic unit 1030 can be implemented using, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field Programmable Gate Array).

[0114] The primary storage device 1040 is a memory device, such as RAM (Random Access Memory), that temporarily stores data used by the arithmetic unit 1030 for various calculations. The secondary storage device 1050 is a storage device where data used by the arithmetic unit 1030 for various calculations and various databases are registered, and can be implemented using ROM (Read Only Memory), HDD (Hard Disk Drive), SSD (Solid State Drive), flash memory, etc. The secondary storage device 1050 may be internal storage or external storage. The secondary storage device 1050 may also be a removable storage medium such as USB (Universal Serial Bus) memory or SD (Secure Digital) memory card. The secondary storage device 1050 may also be cloud storage (online storage), NAS (Network Attached Storage), file server, etc.

[0115] The output I / F 1060 is an interface for transmitting information to be output to output devices 1010, such as displays, projectors, and printers, and is implemented using connectors of standards such as USB (Universal Serial Bus), DVI (Digital Visual Interface), and HDMI (High Definition Multimedia Interface). The input I / F 1070 is an interface for receiving information from various input devices 1020, such as mice, keyboards, keypads, buttons, and scanners, and is implemented using, for example, USB.

[0116] Furthermore, the output interface 1060 and input interface 1070 may be wirelessly connected to the output device 1010 and input device 1020, respectively. In other words, the output device 1010 and input device 1020 may be wireless devices.

[0117] Furthermore, the output device 1010 and the input device 1020 may be integrated as a touch panel. In this case, the output I / F 1060 and the input I / F 1070 may also be integrated as an input / output I / F.

[0118] The input device 1020 may also be a device that reads information from, for example, an optical recording medium such as a CD (Compact Disc), 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.

[0119] The network interface 1080 receives data from other devices via network N and sends it to the computing unit 1030, and also transmits data generated by the computing unit 1030 to other devices via network N.

[0120] The arithmetic unit 1030 controls the output device 1010 and the input device 1020 via the output interface 1060 and the input interface 1070. For example, the arithmetic unit 1030 loads a program from the input device 1020 or the secondary storage device 1050 onto the primary storage device 1040 and executes the loaded program.

[0121] For example, when computer 1000 functions as a server device 100, the arithmetic unit 1030 of computer 1000 realizes the functions of the control unit 130 by executing a program loaded onto the primary storage device 1040. Alternatively, the arithmetic unit 1030 of computer 1000 may load a program obtained from another device via the network interface 1080 onto the primary storage device 1040 and execute the loaded program. Furthermore, the arithmetic unit 1030 of computer 1000 may cooperate with other devices via the network interface 1080 and call and use program functions, data, etc., from other programs on other devices.

[0122] [9. 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.

[0123] 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.

[0124] 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.

[0125] For example, the server device 100 described above may be implemented using multiple server computers, and the configuration can be flexibly changed, such as by calling external platforms via APIs (Application Programming Interfaces) or network computing depending on the function.

[0126] Furthermore, the embodiments and modifications described above can be combined as appropriate, provided that the processing content is not inconsistent.

[0127] 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]

[0128] 1. Information Processing System 10 Terminal devices 100 Server Devices 110 Communications Department 120 Storage section 130 Control Unit 131 Acquisition Department 132 Creation Department 133 Integrated Learning Department 134 Reasoning part 135 Provision Department

Claims

1. An acquisition unit that acquires training data for similar product search and training data for product difference detection, An integrated learning unit performs multitasking learning of the task for searching for similar products and the task for finding product differences, using the learning data for searching for similar products and the learning data for finding product differences. Equipped with, The aforementioned integrated learning unit, In training for similar product search, the product master and product data text are input into the BERT model, and the model is trained on a task of multi-class classification with products as the class for similar product search. In training for product identification, the text of product data is input into a BERT model, and the model is trained to perform a binary classification task for identical products as part of the product identification task. An information processing device characterized by the following:

2. The integrated learning unit inputs training data for similar product search and training data for product difference detection into the same model, and performs training for the similar product search task and the product difference detection task in parallel. The information processing apparatus according to feature 1.

3. The integrated learning unit learns the task of searching for similar products when learning data for searching for similar products is input into the model, and learns the task of finding product differences when learning data for finding product differences is input into the same model. The information processing apparatus according to feature 1.

4. The integrated learning unit, after learning the task of similar product search when learning data for similar product search is input to the model, reuses the parameters of the model to learn the task of product difference detection when learning data for product difference detection is input to the model to which the parameters have been applied. The information processing apparatus according to feature 1.

5. The integrated learning unit, upon inputting training data for similar product search into the model, learns the task for similar product search, adds a task for product error detection to the model, and performs additional learning for product error detection with the model's parameters fixed. The information processing apparatus according to feature 1.

6. The integrated learning unit reuses the parameters of the BERT model trained for similar product search in the BERT model trained for finding differences. In the product difference training, the text of the product data is input into the BERT model that reuses the parameters obtained from the similar product search training, and the unit trains for the task of binary classification of identical products. The information processing apparatus according to feature 1.

7. The integrated learning unit creates a BERT model trained for similar product search, then adds a binary classification layer to the BERT model, fixes the parameters of the BERT model, and performs additional learning for product difference detection. The information processing apparatus according to feature 1.

8. The aforementioned integrated learning unit further performs multitasking learning by adding a task to classify the capacity and number of products. The information processing apparatus according to feature 1.

9. The aforementioned integrated learning unit further performs multitasking learning by adding the task of classifying product brands. The information processing apparatus according to feature 1.

10. An information processing method performed by an information processing device, The acquisition process involves obtaining training data for similar product search and training data for product difference detection. An integrated learning process is performed to perform multitasking learning of a similar product search task and a product difference search task using the learning data for similar product search and the learning data for product difference search. Includes, In the aforementioned integrated learning process, In training for similar product search, the product master and product data text are input into the BERT model, and the model is trained on a task of multi-class classification with products as the class for similar product search. In training for product identification, the text of product data is input into a BERT model, and the model is trained to perform a binary classification task for identical products as part of the product identification task. An information processing method characterized by the following:

11. Procedure for obtaining training data for similar product search and training data for product difference detection, An integrated learning procedure that performs multitasking learning of a similar product search task and a product difference search task using the aforementioned training data for similar product search and the aforementioned training data for product difference search, An information processing program that causes a computer to execute, In the aforementioned integrated learning procedure, In training for similar product search, the product master and product data text are input into the BERT model, and the model is trained on a task of multi-class classification with products as the class for similar product search. In training for product identification, the text of product data is input into a BERT model, and the model is trained to perform a binary classification task for identical products as part of the product identification task. An information processing program characterized by the following features.

Images