Recommendation system, recommendation method, shopping system and shopping method

Abstract

The invention discloses a recommendation system, a recommendation method, a shopping system and a shopping method, and belongs to the technical field of household appliances. The recommendation system comprises a user interaction module which is used for receiving a first input of a user for a wearing and building demand; and the response module is used for responding to the first input and recommending the clothing with the highest matching similarity with the wearing and lapping demand in the clothing database to the user. According to the method and the device, the detailed clothing database is constructed, and the clothing with the highest matching similarity with the wearing demand of the user in the clothing database is recommended to the user in combination with the personal information of the user, the wearing style preference, the purchasing behavior data, the environment information of the user and other multi-dimensional information, so that the accuracy of the clothing recommended to the user is improved.

Description

Technical Field

[0001] This application belongs to the field of home appliance technology, and in particular relates to a recommendation system, recommendation method, shopping system and method. Background Technology

[0002] In today's fast-paced life, people have an increasing demand for fashionable and personalized clothing, but often lack the time and expertise to choose clothes that suit them.

[0003] Clothing recommendation systems in related technologies often fail to adequately consider users' actual needs, resulting in mismatches between recommended clothing and users' actual requirements, leading to poor accuracy. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a recommendation system, recommendation method, shopping system, and method. By constructing a detailed clothing database and combining it with multi-dimensional information such as the user's personal information, clothing style preferences, purchasing behavior data, and the user's environmental information, the system recommends clothing from the database that has the highest similarity to the user's dressing needs, thereby improving the accuracy of clothing recommendations.

[0005] Firstly, this application provides a recommendation system, which includes:

[0006] The user interaction module is used to receive the user's first input regarding their clothing needs. The first input is determined based on the user's personal information, clothing style preferences, purchasing behavior data, and the user's environment information.

[0007] A response module is used to respond to the first input by recommending the clothing in the clothing database that has the highest matching similarity to the dressing requirement to the user. The matching similarity is determined based on the similarity between each piece of clothing in the clothing database and the dressing requirement, as well as the user's acceptance of each recommended piece of clothing in historical recommendation records. The clothing database stores basic information about each piece of clothing.

[0008] According to the recommendation system of this application, by constructing a detailed clothing database and combining it with multi-dimensional information such as the user's personal information, clothing style preferences, purchasing behavior data and the user's environment, the system recommends the clothing in the clothing database that has the highest matching similarity to the user's dressing needs to the user, thereby improving the accuracy of the clothing recommendations.

[0009] According to one embodiment of this application, the response module is further configured to:

[0010] In response to the first input, the large language model is invoked to obtain candidate clothing that meets the dressing requirements;

[0011] Calculate the matching similarity between each garment in the clothing database and the candidate garments;

[0012] The clothing with the highest matching similarity to the candidate clothing in the clothing database is recommended to the user.

[0013] According to the recommendation system of this application, by responding to the user's first input about dressing needs and based on the natural language understanding ability of the large language model, refined data processing and recommendation are performed to obtain candidate clothing that meets the user's dressing needs. The system also queries the clothing database for clothing with the highest matching similarity to the candidate clothing and recommends it to the user, thereby improving the accuracy and personalization of the recommendation system.

[0014] According to one embodiment of this application, after recommending the clothing with the highest matching similarity to the dressing requirement from the clothing database to the user, the response module is further configured to:

[0015] The display interface shows clothing recommended for the user.

[0016] According to the recommendation system of this application, after recommending clothing with the highest similarity to the user's dressing needs, the system uses visualization technology to display the clothing to the user, allowing the user to intuitively view the matching effect of the clothing. This solves the problem of users' difficulty in choosing when faced with a large number of clothing options and improves the user's shopping efficiency.

[0017] According to one embodiment of this application, the user interaction module is further configured to:

[0018] Receive a second input from the user, which indicates the user's acceptance of the recommended clothing;

[0019] Accordingly, the response module is further configured to respond to the second input, and if it is determined that the user does not accept the recommended clothing, recommend other clothing from the clothing database to the user in descending order of the similarity between the other clothing in the clothing database and the dressing requirements.

[0020] According to the recommendation system of this application, users can intuitively view the matching effect of clothing on the display interface. At the same time, the system allows users to interact with the recommendation system by receiving the user's acceptance of the recommended clothing through the user interaction module. The system responds to the user's second input through the response module. If it is determined that the user does not accept the recommended clothing, the system continues to recommend other clothing to the user, thereby optimizing the user's interactive experience and improving the user's browsing and selection experience.

[0021] According to one embodiment of this application, the user interaction module is further configured to:

[0022] Receive third input from the user regarding the aforementioned personal information;

[0023] Accordingly, the response module is also configured to respond to the third input by displaying a first interface for determining the clothing style preference.

[0024] According to one embodiment of this application, the user interaction module is further configured to:

[0025] Receive the user's fourth input regarding their clothing style preference on the first interface;

[0026] Accordingly, the response module is also configured to obtain the clothing style preference in response to the fourth input.

[0027] The recommendation system described in this application collects user personal information, clothing style preferences, purchase history, and real-time environmental data, and combines this information with a clothing database to provide personalized clothing recommendations. By integrating data from different dimensions, the recommendation system is optimized to improve recommendation accuracy and the user's personalized experience.

[0028] Secondly, this application provides a shopping system, which includes a recommendation system as described in the first aspect above, wherein the user interaction module is further configured to receive a fifth input from the user on the display interface regarding a shopping instruction;

[0029] The response module is also configured to respond to the fifth input by displaying a second interface to the user, the second interface being an interface for purchasing clothing recommended to the user.

[0030] The shopping system according to this application integrates e-commerce shopping functions, enabling users to browse, evaluate, and purchase recommended clothing on the same platform. This provides users with a seamless shopping experience, simplifies the shopping process, and may increase users' willingness to purchase goods through the platform.

[0031] According to one embodiment of this application, the user interaction module is further configured to:

[0032] The system receives a sixth input from the user on the second interface, which is used to instruct the user to purchase the recommended clothing.

[0033] Accordingly, the response module is also configured to obtain the user's purchase behavior data in response to the sixth input.

[0034] According to the purchasing system of this application, user purchase behavior data on recommended clothing can provide users' exact preferences and purchase intentions. By recording user purchase behavior, the recommendation system can be further optimized, enabling the recommendation system to recommend clothing to users with higher accuracy.

[0035] Thirdly, this application also provides a recommendation method applied to the recommendation system as described in the first aspect above, comprising:

[0036] The system receives the user's first input regarding their clothing needs, which is determined based on the user's personal information, clothing style preferences, purchasing behavior data, and the user's environment.

[0037] In response to the first input, the clothing in the clothing database with the highest matching similarity to the dressing requirement is recommended to the user. The matching similarity is determined based on the similarity between each piece of clothing in the clothing database and the dressing requirement, as well as the user's acceptance of each recommended piece of clothing in the historical recommendation records. The clothing database stores the basic information of each piece of clothing.

[0038] According to the recommendation method of this application, by constructing a detailed clothing database and combining it with multi-dimensional information such as the user's personal information, clothing style preferences, purchasing behavior data and the user's environment, the clothing in the clothing database with the highest matching similarity to the user's dressing needs is recommended to the user, thereby improving the accuracy of the clothing recommendations.

[0039] Fourthly, this application also provides a shopping method applied to a shopping system as described in the second aspect above, comprising:

[0040] Receive the user's fifth input regarding shopping instructions on the displayed interface;

[0041] In response to the fifth input, a second interface is displayed to the user, which is an interface for purchasing clothing recommended to the user.

[0042] According to the shopping method described in this application, by integrating e-commerce shopping functions, users can browse, evaluate, and purchase recommended clothing on the same platform, providing users with a seamless shopping experience, simplifying the shopping process, and potentially increasing users' willingness to purchase goods through the platform.

[0043] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0044] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0045] Figure 1 This is one of the structural schematic diagrams of the recommendation system provided in the embodiments of this application;

[0046] Figure 2This is a second schematic diagram of the structure of the recommendation system provided in the embodiments of this application;

[0047] Figure 3 This is the third schematic diagram of the structure of the recommendation system provided in the embodiments of this application;

[0048] Figure 4 This is the fourth schematic diagram of the structure of the recommendation system provided in the embodiments of this application;

[0049] Figure 5 This is the fifth schematic diagram of the structure of the recommendation system provided in the embodiments of this application;

[0050] Figure 6 This is the sixth schematic diagram of the structure of the recommendation system provided in the embodiments of this application;

[0051] Figure 7 This is a schematic diagram of the structure of the shopping system provided in the embodiments of this application;

[0052] Figure 8 This is a flowchart illustrating the recommended method provided in the embodiments of this application;

[0053] Figure 9 This is a flowchart illustrating the shopping method provided in the embodiments of this application. Detailed Implementation

[0054] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0055] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0056] Based on the following shortcomings of existing clothing recommendation systems:

[0057] 1. Incomplete data integration: Traditional recommendation systems may not fully integrate user data (such as size and preferences), environmental data (such as weather conditions), and detailed information from clothing databases (such as size charts), resulting in inaccurate recommendations.

[0058] 2. Limited personalization: Existing technologies may not fully utilize personalized data (such as style preferences and purchase history) for personalized outfit recommendations, resulting in a less personalized user experience.

[0059] 3. Insufficient interactive experience: In terms of display and interaction, the existing system may not provide a sufficiently intuitive and flexible interaction method, which limits the user experience and the acquisition of feedback.

[0060] 4. Weak dynamic adjustment capability: Traditional systems may not perform well in dynamically adjusting recommendation strategies based on user behavior (such as purchasing or rejecting recommendations), making it difficult to continuously optimize recommendation performance.

[0061] This application provides a recommendation system that overcomes the aforementioned shortcomings in related technologies and achieves the following effects:

[0062] 1. Improve recommendation accuracy: By building a comprehensive clothing database and combining detailed user information and environmental data, more accurate clothing recommendations can be provided.

[0063] 2. Enhance personalized experience: By deeply analyzing users' style preferences and purchase history, as well as environmental factors such as real-time weather, we provide a more personalized shopping experience and clothing matching recommendations.

[0064] 3. Optimize interaction design: Enhance user interaction experience through visual displays and flexible interaction methods (such as touch screen, voice assistant operation, etc.).

[0065] 4. Dynamically optimize the recommendation algorithm: This application dynamically adjusts the recommendation algorithm by recording user interaction behavior (accepting or rejecting recommended clothing) to achieve higher recommendation accuracy and user satisfaction.

[0066] This application constructs a comprehensive clothing database containing rich information on clothing categories, sizes, colors, and features to ensure users can quickly find satisfactory products during searches. Personal information provided by users during account registration, such as age, gender, and body type, combined with completed clothing style preference questionnaires, allows the system to more accurately understand each user's unique preferences. Furthermore, by integrating users' shopping history (i.e., shopping behavior data) and environmental information (i.e., real-time environmental data such as weather, temperature, and season), the recommendation system can provide the most suitable clothing suggestions at the right time.

[0067] The core of a recommender system is an advanced large language model, capable of handling complex data input, understanding users' personalized needs, and generating clothing and outfit recommendations that match user preferences and the actual environment. The recommendations are then displayed on the user's device through high-quality visualizations. Users can browse different outfits via touchscreen or voice interaction and provide real-time feedback on their preferences, thereby continuously optimizing the system's recommendation algorithm.

[0068] To provide a convenient shopping experience, e-commerce functionality has been integrated, allowing users to directly purchase recommended apparel on partner shopping platforms by clicking the "Buy Now" button. Each purchase serves as feedback input to the system, further refining the personalized recommendation algorithm and ensuring users receive satisfactory shopping suggestions every time. In this way, the recommendation system not only saves users time in selecting clothing but also provides apparel retailers with a tool for precise marketing, achieving a win-win situation.

[0069] The recommendation system, recommendation method, shopping system, and shopping method provided in this application will be described in detail below with reference to the accompanying drawings and through specific embodiments and application scenarios.

[0070] like Figure 1 As shown, the recommendation system includes a user interaction module 100 and a response module 200.

[0071] The user interaction module 100 is used to receive the user's first input regarding clothing needs. The first input is determined based on the user's personal information, clothing style preferences, purchasing behavior data, and the user's environmental information.

[0072] The response module 200 is used to respond to the first input by recommending the clothing in the clothing database that has the highest matching similarity to the dressing requirement to the user. The matching similarity is determined based on the similarity between each piece of clothing in the clothing database and the dressing requirement, as well as the user's acceptance of each recommended piece of clothing in the historical recommendation records. The clothing database stores the basic information of each piece of clothing.

[0073] Understandably, this recommendation system can be applied to smart home devices such as smart refrigerators and smart freezers.

[0074] The user interaction module 100 can specifically be a touch screen provided by the smart home device 300.

[0075] Reference Figure 2The display area provided by the user interaction module 100 can be used to receive the user's first input regarding clothing needs. This first input is determined based on the user's personal information (such as age, gender, height, weight, etc.), clothing style preferences (such as color preferences, pattern selection, occasion requirements, size requirements, etc.), shopping behavior data (such as recently purchased clothing categories, colors, sizes, etc.), and the user's environmental information (such as the weather and season in the user's location).

[0076] This application combines real-time weather information and seasonal change data to provide clothing options suitable for current environmental conditions. This means users will receive practical advice reflecting real-world situations, improving the practicality and comfort of the recommendations.

[0077] The first input can be implemented in at least one of the following ways:

[0078] Firstly, the first input method can be implemented as touch input, including but not limited to click input, swipe input, and press input.

[0079] In this embodiment, receiving the user's first input in the display area can be manifested as receiving the user's touch operation in the display area of ​​the smart home device.

[0080] Secondly, the first input can be implemented as voice command input.

[0081] In this embodiment, receiving the user's first input in the display area can be manifested as receiving the user's voice input from the voice assistant 400 in the display area. For example, a prompt word can be constructed to integrate the user's dressing needs (including the user's personal information (i.e., the user's profile data), clothing style preferences, purchasing behavior data, and the user's environmental information). When the user's voice command input about the prompt word is received, the first input is triggered.

[0082] Of course, in other embodiments, the first input can also be implemented in other forms, including but not limited to motion-sensing gesture input, etc. The specific implementation can be determined according to actual needs, and this application embodiment does not limit it.

[0083] In one example, the prompt could be: "A 28-year-old woman, 168cm tall and weighing 55kg, usually wears size M clothing. She prefers a casual style, cool colors, dislikes patterns, likes cotton materials, and needs breathable, well-fitting everyday casual wear. She is tall and wants to accentuate her legs. She likes designs with pockets and usually wears a watch. She has no particular brand preference, a mid-range budget, and is willing to try new fashion trends. She recently purchased a size M blue casual top. Considering it is currently springtime, the weather in her area is sunny, and the temperature is 23℃, she requests recommendations for suitable clothing."

[0084] Of course, in other embodiments, the first input can also be implemented in other forms, including but not limited to motion-sensing gesture input, etc. The specific implementation can be determined according to actual needs, and this application embodiment does not limit it.

[0085] It should be noted that users access the touch interface provided by smart home devices 300 (such as smart refrigerators, smart freezers, etc.). Figure 3 When registering an account (as shown), users can enter their personal information, including but not limited to age, gender, height, weight, and size.

[0086] This clothing database typically needs to include the following basic information to ensure that clothing can be correctly identified, categorized, recommended, and managed:

[0087] Categories: Tops, pants, skirts, coats, shoes, accessories, etc.

[0088] Size information: XS, S, M, L, XL, etc., as well as specific size data such as bust, waist, hips, length, etc.

[0089] Colors: black, white, red, blue, etc., including different shades and patterns.

[0090] Materials: cotton, wool, silk, denim, synthetic fibers, etc.

[0091] Styles include: casual, business formal, sporty, retro, street, and high-end fashion.

[0092] Suitable seasons: Spring, summer, autumn, and winter.

[0093] Occasions: everyday wear, work, school, parties, formal banquets, outdoor activities, etc.

[0094] Price range: economy, mid-range, high-end, luxury.

[0095] Brand information: including brand awareness, style positioning, target audience, etc.

[0096] Features: comfort, pocket type, closure, garment length, trouser hem design, collar style, sleeve length, etc.

[0097] After a user registers an account through the touch interface of a smart home device, the clothing database will provide a size chart to ensure that the size information entered by the user can match the clothing information in the database.

[0098] This application constructs a clothing database containing multi-dimensional information such as category, size, color, and features to ensure that clothing can be correctly identified, classified, recommended, and managed, achieving comprehensive coverage of clothing attributes.

[0099] When the response module 200 receives the user's first input regarding their outfit requirements, it selects the clothing with the highest similarity to the user's outfit requirements from the clothing database and recommends it to the user.

[0100] The matching similarity score can be determined based on the similarity between each garment in the clothing database and the user's styling needs, as well as the user's acceptance of the recommended garments in historical recommendation records (including whether they accepted or rejected the recommendations). Specifically, the similarity score is set to 1.1 if the user has accepted a recommended garment n times. n If a user rejects a recommended outfit n times, the similarity score is set to 0.9. n .

[0101] For example, a user's specific clothing needs are: "A 28-year-old woman, 168cm tall, weighing 55kg, usually wears size M clothing, prefers a casual style, favors cool colors and dislikes patterns, likes cotton materials, needs suitable everyday casual and breathable clothing, a good fit, is tall, wants to accentuate her legs, likes designs with pockets, usually wears a watch, has no specific brand preference, has a mid-range shopping budget, is willing to try new fashion trends, and recently purchased a size M blue casual top. Considering it is currently springtime, the weather in her area is sunny, and the temperature is 23℃, please return JSON in the specified format."

[0102] The basic information for garment A in the clothing database is as follows: Category: Hoodie; Size: Standard Size "XL"; Specific Dimensions: "Chest": Moderate, "Waist": Moderate, and "Length": Standard; Color: White; Pattern: None; Material: Cotton; Style: Casual; Season: Suitable for Summer; Occasion: Outdoor Activities; Price Range: High-end; Brand Information: No Specific Preference; Features: "Comfort": High, "Pocket Type": Back Pocket, etc.

[0103] The comparison shows that clothing A's category, specific size data, color, pattern, material, brand information, and features all match the user's dressing needs. However, clothing A's size information, season, and price range do not match the user's dressing needs. Therefore, the similarity between clothing A and the user's dressing needs is calculated to be 0.7. Furthermore, based on historical recommendation records, the user has received recommendations for clothing A a total of 10 times. Therefore, the calculated matching similarity is 0.7 × 1.1. 10 .

[0104] According to the recommendation system provided in the embodiments of this application, by constructing a detailed clothing database and combining it with multi-dimensional information such as the user's personal information, clothing style preferences, purchasing behavior data and the user's environment, the system recommends the clothing in the clothing database that has the highest matching similarity to the user's dressing needs to the user, thereby improving the accuracy of the clothing recommended to the user.

[0105] In some embodiments, the response module 200 can also be used for:

[0106] In response to the first input, the large language model is invoked to obtain candidate clothing that meets the dressing requirements;

[0107] Calculate the matching similarity between each garment in the clothing database and the candidate garments;

[0108] The clothing with the highest matching similarity to the candidate clothing in the clothing database is recommended to the user.

[0109] Optionally, when the response module 200 receives the user's first input regarding their outfit requirements in the aforementioned display area, it recommends clothing items from the clothing database that have the highest similarity to the outfit requirements to the user. The specific implementation is as follows:

[0110] The prompt word corresponding to the first input is input into a large language model (such as ChatGLM, Qwen, etc.), and candidate clothing that meets the user's dressing needs is obtained based on the output of the large language model.

[0111] Data cleaning is performed on users' clothing needs, and users' personal information (i.e., user profile data), clothing style preferences, purchase behavior data (i.e., purchase history) and environmental information (i.e. environmental data, such as temperature, weather, season, etc.) are formed into JSON data.

[0112] In one example, clothing needs, including user profile data, clothing style preferences, purchasing behavior data, and environmental information, are converted into JSON data to obtain string-type prompts that the large language model can understand. The resulting JSON data is as follows:

[0113] {

[0114] "User Profile": {

[0115] Age: 28

[0116] Gender: Female

[0117] Height: 168cm

[0118] "Weight": "55kg",

[0119] Size: M

[0120] },

[0121] "Clothing style preference": {

[0122] "Basic Style": "Casual"

[0123] Color preference: Cool colors

[0124] Pattern selection: "None"

[0125] Material preference: "Cotton"

[0126] "Occasion requirement": "Daily leisure"

[0127]

[0128] Based on the aforementioned JSON data, a prompt word is constructed. Specifically, a prompt word is created to integrate the above information and passed to the large language model for data processing and clothing recommendation. The format of the prompt word string should be concise and include all necessary information to ensure that the large language model can understand and respond accordingly. For example:

[0129] The prompt could be: "A 28-year-old woman, 168cm tall and weighing 55kg, usually wears size M clothing. She prefers a casual style, cool colors, and dislikes patterns. She likes cotton materials and needs breathable, well-fitting everyday casual wear. She is tall and wants to accentuate her legs. She likes designs with pockets and usually wears a watch. She has no particular brand preference, a mid-range budget, and is willing to try new fashion trends. She recently purchased a size M blue casual top. Considering it is currently springtime, the weather in her area is sunny, and the temperature is 23℃, please recommend suitable clothing and return JSON data in a specified format."

[0130] Large language models can understand contextual information, such as specific occasions suggested by users (e.g., weekend parties, office work), and adjust recommendation strategies accordingly to ensure that recommended pairings not only match the user's personalized style but are also suitable for specific activities and occasions.

[0131] The prompt words are input into the large language model, and the following is an example of the candidate clothing recommended by the large language model:

[0132]

[0133]

[0134]

[0135] Leveraging the natural language understanding capabilities of the large language model, the system translates user natural language requests into specific queries to the clothing database. The large language module parses the aforementioned JSON data, returns candidate clothing items matching the user's styling needs, and matches them against existing clothing data in the database. It then sorts the candidate clothing items according to their similarity to each item in the database (i.e., the percentage of items in the database that match the candidate clothing item), prioritizing recommendations for the clothing item with the highest similarity.

[0136] This application utilizes large language models to process and analyze user personal information, environmental data, clothing style preferences, and shopping behavior data to generate personalized outfit recommendations. This includes data cleaning, constructing prompts, personalized outfit recommendations, and clothing retrieval and recommendation, demonstrating the capabilities of large language models in parsing and processing natural language requests, and their application in personalized recommendation solutions.

[0137] According to the recommendation system provided in the embodiments of this application, in response to the user's first input about dressing needs, and based on the natural language understanding ability of the large language model, refined data processing and recommendation are performed to obtain candidate clothing that meets the user's dressing needs. The system also queries the clothing database for clothing with the highest matching similarity to the candidate clothing and recommends it to the user, thereby improving the accuracy and personalization of the recommendation system.

[0138] In some embodiments, after recommending the clothing in the clothing database with the highest similarity to the dressing requirement to the user, the response module 200 can also be used to:

[0139] The display interface shows clothing recommended for the user.

[0140] like Figure 3 As shown, the clothing recommended to the user can be displayed on the touchscreen interface of the smart home device 300, or visualized on other devices (e.g., terminals that control the smart home device 300, such as mobile phones, tablets, etc.). Virtual models corresponding to each garment in the clothing database will be used to generate these visual effects, and users can browse the matching effects of different garments through the touchscreen.

[0141] By leveraging virtual models of clothing databases and the visualization capabilities of smart home devices, users can intuitively see how recommended outfits will look. This not only enhances the user experience but also helps users better understand and evaluate the recommended clothing items.

[0142] It should be noted that this application can also integrate augmented reality (AR) try-on functionality as an alternative to visual display. AR technology can be used to allow users to see themselves wearing recommended clothing in their own living environment through a smart device's camera. This can increase user engagement and provide a more realistic try-on experience.

[0143] This application can also integrate a smart wardrobe management system, allowing users to upload their own clothing to the platform. The recommendation system not only suggests purchasing new clothes but also provides outfit suggestions based on the user's existing clothing.

[0144] In addition, this application can also integrate a fashion expert system, allowing experts to regularly update advanced knowledge such as fashion trends and color theory, thereby improving the quality of the recommendation system.

[0145] According to the recommendation system provided in the embodiments of this application, after recommending the clothing with the highest similarity to the user's dressing needs, the system uses visualization technology to display the clothing to the user, allowing the user to intuitively view the matching effect of the clothing. This solves the problem of users having difficulty choosing when faced with a large number of clothing options and improves the user's shopping efficiency.

[0146] In some embodiments, the user interaction module 100 can also be used for:

[0147] Receive a second input from the user, which indicates the user's acceptance of the recommended clothing;

[0148] Accordingly, the response module is further configured to respond to the second input, and if it is determined that the user does not accept the recommended clothing, recommend other clothing from the clothing database to the user in descending order of the similarity between the other clothing in the clothing database and the dressing requirements.

[0149] Optionally, the user interaction module 100 can also be used to receive a second input from the user, which can specifically be the user's acceptance of the recommended clothing, including accepting the recommended clothing or rejecting the recommended clothing (i.e., not being interested in the recommended clothing). The user can interact through the touch screen to change different clothing options.

[0150] The second input can be implemented in at least one of the following ways:

[0151] Firstly, the second input method can be implemented as touch input, including but not limited to click input, swipe input, and press input.

[0152] In this embodiment, receiving the user's second input on the display area can manifest as receiving the user's touch operation on the display area of ​​the smart home device. For example, such as... Figure 4 As shown, this second input is triggered when the user taps the "Next" or "Change" icon control in the display area. Alternatively, it is triggered when the user taps the "OK" icon control in the display area.

[0153] Secondly, the second input can be implemented as voice command input.

[0154] In this embodiment, receiving the user's second input in the display area can manifest as receiving the user's voice input from the voice assistant 400 in the display area. For example... Figure 4 As shown, for example, the second input is triggered when the user says "next" or "change one" to the voice assistant 400 using natural language. Or, the second input is triggered when the user says "OK" or "okay" to the voice assistant 400 using natural language.

[0155] Of course, in other embodiments, the second input can also be implemented in other forms, including but not limited to motion-sensing gesture input, which can be determined according to actual needs. This application embodiment does not limit this.

[0156] When the response module 200 receives the user's second input, if it determines that the user does not accept the recommended clothing, it will recommend the clothing in the clothing database to the user in descending order of the matching similarity between the remaining clothing in the clothing database and the user's dressing needs (i.e., the candidate clothing recommended by the large language model), and display them to the user on the display interface.

[0157] It's important to note that each interaction between a user and the recommendation system is considered a learning opportunity. The system records the user's choices (accepting or rejecting recommended clothing items) and uses this information to dynamically adjust the recommendation algorithm to more accurately predict user preferences. This continuous learning process ensures that the recommendation system becomes increasingly intelligent and personalized over time.

[0158] Unlike clothing recommendation systems in related technologies, which cannot effectively learn from user behavior to optimize future recommendations, this application continuously adjusts its recommendation algorithm by tracking and analyzing user interactions (accepting or rejecting recommended outfits), purchasing behavior, etc. This not only reflects changes in user preferences in real time but also makes the system's recommendations increasingly accurate and personalized over time.

[0159] Recommended combinations are visually displayed on the touchscreens of smart home devices such as smart refrigerators and smart freezers, or on other terminals that control these smart home devices, allowing users to interact with the devices via the touchscreen, optimize their choices, and enhance the user's interactive experience.

[0160] The recommendation system provided in this application embodiment allows users to intuitively view the matching effect of clothing on the display interface. At the same time, the system enables users to interact with the recommendation system by receiving the user's acceptance of the recommended clothing through the user interaction module. The system responds to the user's second input through the response module. If it is determined that the user does not accept the recommended clothing, the system continues to recommend other clothing to the user, thereby optimizing the user's interactive experience and improving the user's browsing and selection experience.

[0161] In some embodiments, the user interaction module 100 can also be used for:

[0162] Receive third input from the user regarding the aforementioned personal information;

[0163] Accordingly, the response module is also configured to respond to the third input by displaying a first interface for determining the clothing style preference.

[0164] like Figure 5 As shown, the touch interface provided by the user interaction module 100 can be used to receive third-party input from the user regarding personal information, which includes, but is not limited to, age, gender, height, weight, and size.

[0165] The third input can be implemented in at least one of the following ways:

[0166] Firstly, the third input method can be implemented as touch input, including but not limited to click input, swipe input, and press input.

[0167] In this embodiment, receiving a third input from the user on the touch interface can be manifested as receiving a touch operation from the user on the touch interface of a smart home device.

[0168] Secondly, the third input can be implemented as voice command input.

[0169] In this embodiment, receiving the user's third input on the touch interface can be manifested as receiving the user's voice input from the voice assistant 400 in the display area.

[0170] Of course, in other embodiments, the third input can also be implemented in other forms, including but not limited to motion-sensing gesture input, etc. The specific implementation can be determined according to actual needs, and this application embodiment does not limit it.

[0171] When the response module 200 receives a third input from the user regarding personal information, it displays a first interface for the user to set clothing style preferences.

[0172] In some embodiments, the user interaction module is further configured to:

[0173] Receive the user's fourth input regarding their clothing style preference on the first interface;

[0174] Accordingly, the response module is also configured to obtain the clothing style preference in response to the fourth input.

[0175] like Figure 6 As shown, the first interface provided by the user interaction module 100 can be used to receive a fourth input from the user regarding their clothing style preferences. These preferences can specifically include basic style preferences, colors and patterns, materials and textures, occasions and functions, size and comfort, personal characteristics, accessories and details, brands and prices, and fashion trends. Users can complete the clothing style preference questionnaire provided on the first interface as needed. The system will record the user's answers and determine their clothing style preferences based on the recorded responses. The clothing style preference questionnaire aims to better understand the user's clothing preferences for providing suggestions and customizing a personalized shopping experience. An example of a clothing style preference questionnaire is shown below:

[0176] 1. Basic style preferences:

[0177] What type of clothing do you usually prefer (formal, casual, sporty, cutting-edge fashion, etc.)?

[0178] Which clothing style do you prefer (simple, retro, street style, etc.)?

[0179] Which type of clothing cut do you usually prefer (fitted, loose, straight, layered, etc.)?

[0180] 2. Colors and patterns:

[0181] What are your preferred colors (e.g., warm tones, cool tones, bright colors, neutral colors, etc.)?

[0182] Which pattern do you usually choose (stripes, checks, abstract, animal prints, plain patterns, etc.)?

[0183] 3. Materials and textures:

[0184] Which material do you prefer (cotton, silk, wool, synthetic fibers, etc.)?

[0185] What are your preferences regarding the texture of clothing (smooth, rough, glossy, matte, etc.)?

[0186] 4. Occasion and function:

[0187] What occasions (work, leisure, sports, dinner, etc.) do you need suitable clothing for?

[0188] Do you need to consider any special features (such as waterproof, wrinkle-resistant, elastic, breathable, etc.)?

[0189] 5. Size and comfort:

[0190] What size do you usually wear (e.g., S, M, L, XL, XXL, etc.)?

[0191] What are the requirements for the fit of clothes (e.g., loose, moderate, tight, etc.)?

[0192] 6. Personal characteristics:

[0193] Do you have any personal body type characteristics to consider (such as tall, short, slender, muscular, etc.)?

[0194] Are there any body parts you wish to highlight or conceal with clothing (such as calves, ankles, arms, etc.)?

[0195] 7. Accessories and details:

[0196] What details do you like in clothing (such as pockets, buttons, embroidery, zippers, etc.)?

[0197] What kind of accessories (hats, scarves, jewelry, watches, etc.) do you usually wear?

[0198] 8. Brand and Price:

[0199] Do you have a preference for specific brands (such as Li-Ning, Semir, etc.)?

[0200] What is your shopping budget range?

[0201] 9. Fashion Trends:

[0202] Are you willing to try the latest fashion trends (e.g., no, so-so, yes, etc.)?

[0203] The fourth input can be implemented in at least one of the following ways:

[0204] Firstly, the fourth input method can be implemented as touch input, including but not limited to click input, swipe input, and press input.

[0205] In this embodiment, receiving the user's fourth input on the first interface can be manifested as receiving the user's touch operation on the first interface of the smart home device.

[0206] Secondly, the fourth input can be implemented as voice command input.

[0207] In this embodiment, receiving the user's fourth input on the first interface can be manifested as receiving the user's voice input from the voice assistant 400 on the first interface.

[0208] Of course, in other embodiments, the fourth input can also be implemented in other forms, including but not limited to motion-sensing gesture input, etc. The specific implementation can be determined according to actual needs, and this application embodiment does not limit it.

[0209] The recommendation system provided in this application collects user personal information, clothing style preferences, purchase history, and real-time environmental data, and combines this information with a clothing database to provide personalized clothing recommendations. By integrating data from different dimensions, the recommendation system is optimized, improving recommendation accuracy and the user's personalized experience.

[0210] This application synchronizes purchase history from a shopping platform, matching user purchase data with clothing entries in a clothing database to more accurately analyze user style preferences. It also automatically obtains real-time weather information for the user's location, taking seasonal changes into account, to influence clothing recommendations.

[0211] This application also provides a shopping system, which includes the recommendation system described above, wherein...

[0212] The user interaction module 100 is also used to receive a fifth input from the user on the display interface regarding shopping instructions;

[0213] The response module 200 is also configured to respond to the fifth input by displaying a second interface to the user, the second interface being an interface for purchasing clothing recommended to the user.

[0214] like Figure 7 As shown, the display interface provided by the user interaction module 100 (i.e., the recommended clothing matching display interface) can also be used to receive the user's fifth input for shopping instructions.

[0215] The fifth input can be implemented in at least one of the following ways:

[0216] Firstly, the fifth input method can be implemented as touch input, including but not limited to click input, swipe input, and press input.

[0217] In this embodiment, receiving the fifth input from the user on the display interface can manifest as receiving the user's touch operation on the display interface of the smart home device. For example, the recommended clothing matching display interface (i.e., the display interface) may provide a "Buy Now" icon control. When the user clicks the "Buy Now" icon control on the display interface, the fifth input is triggered.

[0218] Secondly, the fifth input method can be implemented as voice command input.

[0219] In this embodiment, receiving the fifth input from the user on the display interface can be manifested as receiving the user's voice input from the voice assistant 400 on the display interface. For example, this fifth input is triggered when the user says natural expressions such as "buy now" or "buy" to the voice assistant 400.

[0220] Of course, in other embodiments, the fifth input can also be implemented in other forms, including but not limited to motion-sensing gesture input, which can be determined according to actual needs. This application embodiment does not limit this.

[0221] When the response module 200 receives the user's fifth input, it displays a second interface to the user, allowing the user to seamlessly purchase recommended clothing. This second interface refers to the interface for purchasing clothing recommended to the user, or the purchase interface.

[0222] The shopping system provided in the embodiments of this application integrates e-commerce shopping functions, enabling users to browse, evaluate, and purchase recommended clothing on the same platform. This provides users with a seamless shopping experience, simplifies the shopping process, and may increase users' willingness to purchase goods through the platform.

[0223] In some embodiments, the user interaction module 100 can also be used for:

[0224] The system receives a sixth input from the user on the second interface, which is used to instruct the user to purchase the recommended clothing.

[0225] Accordingly, the response module is also configured to obtain the user's purchase behavior data in response to the sixth input.

[0226] Optionally, the second interface provided by the user interaction module 100 can be used to receive a sixth input from the user, wherein the sixth input can be used to instruct the user to purchase the recommended clothing.

[0227] The sixth input can be implemented in at least one of the following ways:

[0228] Firstly, the sixth input method can be implemented as touch input, including but not limited to click input, swipe input, and press input.

[0229] In this embodiment, receiving the user's sixth input on the second interface can be manifested as receiving the user's touch operation on the second interface of the smart home device.

[0230] Secondly, the sixth input can be implemented as voice command input.

[0231] In this embodiment, receiving the user's sixth input on the second interface can be manifested as receiving the user's voice input from the voice assistant 400 on the second interface.

[0232] Of course, in other embodiments, the sixth input can also be implemented in other forms, including but not limited to motion-sensing gesture input, which can be determined according to actual needs. This application embodiment does not limit this.

[0233] When the response module 200 receives the user's sixth input, it obtains the user's purchase behavior data based on the user's purchase behavior.

[0234] It's important to note that every purchase a user makes is recorded and analyzed by the recommendation system to further optimize the personalized recommendation algorithm. Purchase behavior data provides direct feedback, revealing a user's precise preferences and purchase intentions, enabling the recommendation system to adjust future recommendations with greater accuracy, thus helping to improve user satisfaction and purchase conversion rates.

[0235] Unlike clothing recommendation systems in related technologies, where users typically need to go through a complex purchase process after finding suitable clothing, this application simplifies the shopping process by integrating e-commerce functions, such as a "Buy Now" button. Users can directly enter the purchase stage from the recommendation interface, greatly improving the convenience of shopping.

[0236] Through the aforementioned e-commerce integration functions, the shopping system is not only an intelligent clothing matching and recommendation tool, but also a comprehensive online shopping assistant, providing a one-stop service from discovery and recommendation to purchase and feedback, greatly enhancing the user experience.

[0237] According to the purchasing system provided in the embodiments of this application, the user's purchasing behavior data on recommended clothing can provide the user's exact preferences and purchasing intentions. By recording the user's purchasing behavior, the recommendation system can be further optimized, enabling the recommendation system to recommend clothing to users with higher accuracy.

[0238] This application also provides a recommendation method. The method is applied to the recommendation system described above.

[0239] like Figure 8 The recommended method includes steps 110 and 120.

[0240] Step 110: Receive the user's first input regarding clothing needs. The first input is determined based on the user's personal information, clothing style preferences, purchasing behavior data, and the user's environment.

[0241] Step 120: In response to the first input, recommend the clothing in the clothing database with the highest matching similarity to the dressing requirement to the user. The matching similarity is determined based on the similarity between each piece of clothing in the clothing database and the dressing requirement, as well as the user's acceptance of each recommended piece of clothing in the historical recommendation records. The clothing database stores the basic information of each piece of clothing.

[0242] According to the recommendation method of this application embodiment, by constructing a detailed clothing database and combining it with multi-dimensional information such as the user's personal information, clothing style preferences, purchasing behavior data and the user's environment, the clothing in the clothing database with the highest matching similarity to the user's dressing needs is recommended to the user, thereby improving the accuracy of the clothing recommended to the user.

[0243] This application also provides a shopping method. The method is applied to the shopping system described above.

[0244] like Figure 9 As shown, the shopping method includes steps 210 and 220.

[0245] Step 210: Receive the user's fifth input regarding shopping instructions on the displayed interface;

[0246] Step 220: In response to the fifth input, a second interface is displayed to the user, the second interface being an interface for purchasing clothing recommended to the user.

[0247] According to the shopping method provided in the embodiments of this application, by integrating e-commerce shopping functions, users can browse, evaluate and purchase recommended clothing on the same platform, providing users with a seamless shopping experience, simplifying the shopping process, and potentially increasing users' willingness to purchase goods through the platform.

[0248] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0249] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0250] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0251] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0252] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A recommendation system, characterized in that, include: The user interaction module is used to receive the user's first input regarding their clothing needs. The first input is determined based on the user's personal information, clothing style preferences, purchasing behavior data, and the user's environment information. A response module is used to respond to the first input by recommending the clothing in the clothing database that has the highest matching similarity to the dressing requirement to the user. The matching similarity is determined based on the similarity between each piece of clothing in the clothing database and the dressing requirement, as well as the user's acceptance of each recommended piece of clothing in historical recommendation records. The clothing database stores basic information about each piece of clothing.

2. The recommendation system according to claim 1, characterized in that, The response module is also used for: In response to the first input, the large language model is invoked to obtain candidate clothing that meets the dressing requirements; Calculate the matching similarity between each garment in the clothing database and the candidate garments; The clothing with the highest matching similarity to the candidate clothing in the clothing database is recommended to the user.

3. The recommendation system according to claim 1 or 2, characterized in that, After recommending the clothing items in the clothing database that have the highest similarity to the user's dressing needs, the response module is further configured to: The display interface shows clothing recommended for the user.

4. The recommendation system according to claim 3, characterized in that, The user interaction module is also used for: Receive a second input from the user, which indicates the user's acceptance of the recommended clothing; Accordingly, the response module is further configured to respond to the second input, and if it is determined that the user does not accept the recommended clothing, recommend other clothing from the clothing database to the user in descending order of the similarity between the other clothing in the clothing database and the dressing requirements.

5. The recommendation system according to claim 3, characterized in that, The user interaction module is also used for: Receive third input from the user regarding the aforementioned personal information; Accordingly, the response module is also configured to respond to the third input by displaying a first interface for determining the clothing style preference.

6. The recommendation system according to claim 5, characterized in that, The user interaction module is also used for: Receive the user's fourth input regarding their clothing style preference on the first interface; Accordingly, the response module is also configured to obtain the clothing style preference in response to the fourth input.

7. A shopping system, characterized in that, Including the recommendation system as described in any one of claims 1-6, The user interaction module is also used to receive a fifth input from the user regarding shopping instructions on the display interface; The response module is also configured to respond to the fifth input by displaying a second interface to the user, the second interface being an interface for purchasing clothing recommended to the user.

8. The shopping system according to claim 7, characterized in that, The user interaction module is also used for: The system receives a sixth input from the user on the second interface, which is used to instruct the user to purchase the recommended clothing. Accordingly, the response module is also configured to obtain the user's purchase behavior data in response to the sixth input.

9. A recommended method, characterized in that, Applied to the recommendation system as described in any one of claims 1-6, comprising: The system receives the user's first input regarding their clothing needs, which is determined based on the user's personal information, clothing style preferences, purchasing behavior data, and the user's environment. In response to the first input, the clothing in the clothing database with the highest matching similarity to the dressing requirement is recommended to the user. The matching similarity is determined based on the similarity between each piece of clothing in the clothing database and the dressing requirement, as well as the user's acceptance of each recommended piece of clothing in the historical recommendation records. The clothing database stores the basic information of each piece of clothing.

10. A shopping method, characterized in that, Applied to the shopping system as described in any one of claims 7-8, comprising: Receive the user's fifth input regarding shopping instructions on the displayed interface; In response to the fifth input, a second interface is displayed to the user, which is an interface for purchasing clothing recommended to the user.

Images