Commodity recommendation method, computer readable storage medium, electronic device

Abstract

Embodiments of the present application disclose a commodity recommendation method, a computer readable storage medium and an electronic device. The method comprises: receiving existing food material information uploaded by a user and a cooking intention expressed by the user; analyzing the existing food material information and the cooking intention by an artificial intelligence (AI) model, determining a target dish to be made, and generating making method information of the target dish and missing food material information of the target dish according to knowledge information related to making the target dish; determining recommendation commodity information from salable commodities provided by a commodity information service system according to the missing food material information; and providing the making method information of the target dish and the recommendation commodity information to a client. The method can ensure the fitting degree between recipe recommendation and commodity recommendation, better fit the daily consumption habits of the user, and improve the shopping experience of the user.

Description

Technical Field

[0001] This application relates to the field of commodity information service technology, and in particular to a commodity recommendation method, a computer-readable storage medium, an electronic device, and a computer program product. Background Technology

[0002] To optimize the user shopping experience, when consumers are selecting goods online, the product information service system can provide information related to the preparation and processing methods of the products. For example, after a user searches for "tomato," the search results page or product details page can display the preparation methods (recipes) for dishes such as scrambled eggs with tomatoes. Simultaneously, other ingredients needed to prepare the dish can be displayed as recommended products. Users can then select both the searched product and the recommended products as their desired items and submit an online order. This approach helps improve the conversion rate of searched products, increases the exposure of recommended products, and provides users with supplementary recipe guidance, thus enhancing the overall user shopping experience.

[0003] How to further optimize the solution to ensure that the recipes provided by the product information service system better meet users' shopping needs has become a technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] This application provides a product recommendation method and apparatus, a computer-readable storage medium, an electronic device, and a computer program product. Based on the user's uploaded existing ingredient information and cooking intentions, it recommends recipes and products, which can ensure the compatibility between recipe recommendations and product recommendations, better fit the user's daily consumption habits, and improve the user's shopping experience.

[0005] This application provides the following solution: A product recommendation method, the method comprising: Receive information about existing ingredients uploaded by users, as well as their cooking intentions; The existing ingredient information and cooking intention are analyzed by an artificial intelligence (AI) model to determine the target dish to be made. Based on the knowledge information related to making the target dish, the preparation method information of the target dish and the ingredient information that is still missing in making the target dish are generated. Based on the missing ingredient information, recommended product information is determined from the available products provided by the product information service system; Provide the client with information on how the target dish is prepared, as well as information on recommended products.

[0006] The existing ingredient information and the cooking intention are expressed in a multimodal manner in the form of images and / or text and / or voice. The AI ​​model includes a multimodal language model, or includes multiple unimodal language models adapted to the expression of the existing ingredient information and the cooking intention.

[0007] The existing ingredient information refers to images obtained by taking pictures of the ingredients that the user has stored, and the cooking intention is expressed in the form of text and / or voice.

[0008] The information regarding the preparation method for generating the target dish includes: A health label representing the user's dietary preferences is obtained and used as a prompt word for the AI ​​model. The AI ​​model then generates information on the preparation method of the target dish based on the health label.

[0009] The knowledge information includes knowledge related to the substitution relationships of different ingredients, and the information on the preparation method for generating the target dish includes: Based on the knowledge related to the substitution relationship, it is determined that there is a first ingredient that can be substituted in the target dish, and the existing ingredient information includes substitute ingredients for the first ingredient. The preparation method information of the target dish is generated using the substitute ingredients.

[0010] The method further includes, after providing the client with the preparation method information of the target dish, the following: Receive correction information submitted by the user regarding the preparation method information, the correction information including information related to the user's taste preferences and / or kitchenware usage preferences; The correction information is used as a prompt word for the AI ​​model, which then adjusts the preparation method information of the target dish based on the correction information.

[0011] This includes receiving the user's expressed cooking intentions, including: The existing ingredient information is used as the main ingredient in the dish, and the AI ​​model is used to identify the user's cooking intention. The identified cooking intention is provided to the user for confirmation, and the cooking intention confirmed by the user is taken as the cooking intention expressed by the user.

[0012] If at least two cooking intentions are identified for the same main ingredient, the method further includes: For the at least two cooking intentions, generate dish characteristic tags and provide them to the user for confirmation of the cooking intentions; the dish characteristic tags are used to represent the characteristic information of the dish related to the cooking intention.

[0013] The method further includes: Provide the missing ingredient information to the user for confirmation; The recommended product information is determined based on the ingredients selected by the user from the missing ingredient information.

[0014] A product recommendation method, the method comprising: The client receives information about existing ingredients uploaded by the user and their cooking intentions, and sends it to the server. The server then uses an artificial intelligence (AI) model to analyze the information and intentions to determine the target dish to be made. Based on knowledge related to making the target dish, the server generates information about the preparation method and any missing ingredients. Based on the missing ingredient information, the server determines recommended products from the available products provided by the product information service system. The system receives and displays the preparation method information of the target dish and the recommended product information sent by the server.

[0015] The client establishes a session connection with the server via intelligent dialogue and provides a session interface to obtain information about existing ingredients and the cooking intention, and to display information about the preparation method of the target dish and recommended products through the session interface.

[0016] A product recommendation device, applied on a server side, the device comprising: The information receiving unit is used to receive information about existing ingredients uploaded by the user and the cooking intention expressed. The information generation unit is used to analyze the existing ingredient information and cooking intention through an artificial intelligence (AI) model to determine the target dish to be made, and generate the cooking method information of the target dish and the ingredient information that is still missing in making the target dish based on the knowledge information related to making the target dish. The recommended product determination unit is used to determine recommended product information from the available products provided by the product information service system based on the missing ingredient information; The information providing unit is used to provide the client with information on the preparation method of the target dish and the recommended product information.

[0017] A product recommendation device, applied to a client, the device comprising: The information sending unit is used to receive existing ingredient information and the cooking intention expressed by the user, and send them to the server. The server then uses an artificial intelligence (AI) model to analyze the existing ingredient information and cooking intention to determine the target dish to be made. Based on the knowledge information related to making the target dish, the server generates information on the preparation method of the target dish and information on any missing ingredients. Based on the missing ingredient information, the server determines recommended product information from the available products provided by the product information service system. The information receiving unit is used to receive and display the preparation method information of the target dish and the recommended product information sent by the server.

[0018] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of any of the preceding methods.

[0019] An electronic device, comprising: One or more processors; and A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of any of the preceding methods.

[0020] A computer program product includes a computer program / computer executable instructions that, when executed by a processor in an electronic device, implement the steps of any of the preceding methods.

[0021] According to the specific embodiments provided in this application, the following technical effects are disclosed: In this embodiment, an AI model can intelligently analyze the user's uploaded ingredient information and expressed cooking intentions to determine the target dish the user wishes to make. Furthermore, based on relevant knowledge about the target dish, it generates information on the preparation method and any missing ingredients. The preparation method information is provided to the user for recipe recommendation, and product recommendations are provided for the missing ingredients, ensuring a good match between recipe and product recommendations. Moreover, this application utilizes the user's existing ingredient reserves as a technological foundation. By consuming existing ingredients, it drives the exposure and sales conversion of products containing missing ingredients. This product recommendation scheme is more targeted and better aligns with the daily consumption habits of most users, contributing to an improved user shopping experience.

[0022] Of course, any product implementing this application does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the product recommendation system provided in the embodiments of this application; Figure 2 This is a flowchart of the product recommendation method provided in the embodiments of this application; Figure 3 This is a schematic diagram of a product recommendation device provided in an embodiment of this application; Figure 4 This is a schematic diagram of another product recommendation device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. 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.

[0026] In the existing product recommendation scheme, consumers can enter the keyword "tomato" into the search results. The relevant page in the product information service system can then display the preparation methods for dishes such as scrambled eggs with tomatoes—i.e., recipes. Simultaneously, other ingredients needed to prepare the dish, such as "eggs," can be selected as recommended products for promotion and exposure. By recommending recipes, both the searched product and the other ingredients required for the dish can be simultaneously exposed, thereby increasing the conversion rate of sales.

[0027] This application improves the recipe recommendation method. Instead of recommending recipes and exposing products based on the search results of products the user wants to buy, it improves the method by recommending recipes based on the user's existing ingredients and then recommending and exposing products for ingredients missing from the recipe. This application focuses on using the user's existing ingredients to recommend recipes and products for missing ingredients. By consuming existing ingredients, it drives the exposure and conversion of products for missing ingredients. The recommendation scheme is more targeted and better aligns with the daily consumption habits of most users, thus improving the user shopping experience.

[0028] From a system architecture perspective, the product recommendation system in this application embodiment can be as follows: Figure 1 As shown, it includes: client and server.

[0029] The client can be deployed on the user's terminal device to receive the information of existing ingredients uploaded by the user and the cooking intention expressed. It then sends this information to the server for intelligent recipe and product recommendations. The client also receives the preparation method information and recommended product information of the target dish returned by the server and displays them so that the user can purchase missing ingredients based on the recommended product information and process the target dish based on the preparation method information.

[0030] The server can be deployed on a cloud server. After receiving the existing ingredient information and cooking intention, it uses an AI model to intelligently analyze the information and determine the target dish that the user wants to make. Then, based on the knowledge information of the dish, it generates the cooking method information of the target dish and the information of the ingredients that are still missing in making the target dish. Based on the missing ingredient information, it determines the corresponding recommended product information, thus realizing the recipe recommendation and product recommendation of this application with the purpose of consuming the user's existing ingredients.

[0031] The implementation process of the product recommendation method in this application embodiment will be explained below with specific examples. See [link to relevant documentation]. Figure 2 The flowchart shown may include: S201: The client receives the information on existing ingredients uploaded by the user and the cooking intention expressed, and sends it to the server.

[0032] In this embodiment, existing ingredient information and cooking intentions can be expressed in multimodal form using images and / or text and / or voice. Users can upload existing ingredient information and cooking intentions through various interactive methods. Existing ingredients refer to ingredients that the user currently has in stock and that can be used in the current dish preparation process. This application can use existing ingredients as the ingredients that the user wishes to consume during the current dish preparation process to provide targeted recipe and product recommendations.

[0033] For existing ingredient information, users can take pictures of the ingredients they have stored to obtain images containing the ingredients and upload them to the server as existing ingredient information; alternatively, users can input descriptions of the ingredients they have stored via voice or text and upload these descriptions to the server as existing ingredient information. The descriptions should at least include the product identifier of the existing ingredients.

[0034] Regarding cooking intent, the system can obtain an image of the dish the user wishes to make and upload it to the server as the cooking intent; alternatively, the user can express their cooking intent via voice or text and upload it to the server. As an example, the user can proactively express their cooking intent during the interaction; alternatively, if it's determined that the user hasn't uploaded a cooking intent, the system can guide the user to express it through dialogue; or, if it's determined that the user hasn't uploaded a cooking intent, the system can also identify the cooking intent based on the user's uploaded ingredient information and provide it to the user for confirmation.

[0035] In practical applications, it is common to encounter situations where users are at a loss when faced with ingredients stored at home. If users do not have a clear cooking intention for the ingredients they already have, they are more likely to seek cooking inspiration from the server by providing information about the ingredients they already have through this application solution.

[0036] Furthermore, this application can classify the ingredients required for cooking dishes into two categories: main ingredients and auxiliary ingredients. In practical applications, the same ingredient may belong to different categories in different dishes. Taking chili peppers as an example, chili peppers are an ingredient required for the dish "Stir-fried Pork with Chili Peppers" and belong to the category of processed main ingredients; at the same time, they are also an ingredient required for the dish "Fish-flavored Shredded Pork" but belong to the category of processed auxiliary ingredients.

[0037] Correspondingly, when the user does not explicitly provide a cooking intention, the server can use existing ingredients as a basic requirement. Based on the constraint that the user's existing ingredients are the main ingredients for processing, the server can identify the cooking intention. That is, it uses the existing ingredients as the main ingredients for making the dish to identify the user's cooking intention. For example, if the user uploads information about existing ingredients including "potatoes" and "beef," the corresponding identified cooking intentions may include "stewed beef with potatoes," "curry beef with potatoes," "dry pot beef with potatoes," etc. The intelligently identified cooking intentions can be provided to the user through the client for confirmation, and the user's confirmed "curry beef with potatoes" can be taken as the cooking intention expressed by the user.

[0038] Optionally, if at least two cooking intentions are identified for the same main ingredient, the server can also generate dish characteristic tags for these intentions, providing them to the user for confirmation. These dish characteristic tags represent the characteristics of the dish related to the cooking intention, such as nutritional balance, healthy weight loss, and quick and easy preparation. Users can confirm their cooking intentions based on their specific needs and the dish characteristic tags. For example, in the example above, the cooking intention "stewed beef with potatoes" is labeled "healthy and weight loss friendly," the cooking intention "curry beef with potatoes" is labeled "quick and easy, low difficulty to prepare," and the cooking intention "dry pot beef with potatoes" is labeled "rich flavor."

[0039] In this embodiment, the client can establish a session connection with the server via intelligent dialogue and provide a session interface to obtain existing ingredient information and cooking intentions. Taking uploading existing ingredient information as an example, the client can provide an image capture function through the session interface. After the consumer user activates the image capture function, they can take real-time pictures of the ingredients stored in the refrigerator and send them to the server as part of the dialogue. Alternatively, the client can provide an image upload function through the session interface. The user can activate the image upload function to send locally taken pictures of existing ingredients to the server as part of the dialogue.

[0040] S202: The server uses an artificial intelligence (AI) model to analyze existing ingredient information and cooking intentions, determines the target dish to be made, and generates information on the preparation method of the target dish, as well as information on ingredients that are still missing in the preparation of the target dish, based on knowledge related to the preparation of the target dish.

[0041] In this embodiment, the server can be represented as an intelligent agent. After receiving the user's uploaded information about existing ingredients and their expressed cooking intention, the agent can use an AI model to analyze the information and intention to determine the target dish to be made. For example, if the user uploads information about existing ingredients including "potatoes" and "beef" and expresses their cooking intention as "want to make curry," the target dish determined by the AI ​​model is curry potato and beef.

[0042] Specifically, an intelligent agent is an entity capable of perceiving its environment, making decisions, and executing actions. It typically consists of multiple key modules, including, for example, a "brain" (such as a generative language model or artificial intelligence algorithm), a perception module, a memory and retrieval module, and an action execution module. AI models, such as generative language models or multimodal models, are deep learning models trained on massive amounts of data. They excel at pattern recognition and generation, understanding natural language, images, and other information, and generating logically consistent text, code, and images. The intelligent agent and the AI ​​model have a complementary relationship. The AI ​​model acts as the "core capability engine" or "core brain" of the intelligent agent, providing it with powerful cognitive reasoning abilities; while the intelligent agent acts as the "system carrier" or "capability extension" of the AI ​​model, autonomously completing specific tasks using the AI ​​model's capabilities.

[0043] In this embodiment, both existing ingredient information and cooking intention can be expressed multimodally in the form of images and / or text and / or voice. The corresponding AI model can include a Multimodal Large Language Model (MLLM), which is an artificial intelligence model capable of simultaneously processing and understanding multiple types of data (such as text, images, audio, video, etc.). Alternatively, the AI ​​model can include multiple unimodal language models adapted to the expression forms of existing ingredient information and cooking intention. For example, it can include a text language model and a visual language model. The visual language model can identify existing ingredient information in the image, and the text language model can parse the user's cooking intention.

[0044] By identifying the target dish to be prepared using an AI model, information on the preparation method and any missing ingredients can be generated based on relevant knowledge about the target dish.

[0045] In this embodiment, the knowledge information related to dish preparation can be general knowledge related to dish preparation obtained by the AI ​​model based on massive amounts of data. This general knowledge can be used to generate information on the preparation method of the target dish and identify missing ingredient information. Optionally, to improve the accuracy of the model's output, some external professional knowledge related to dish preparation can also be provided. The AI ​​model generates recipes based on general knowledge and external professional knowledge and identifies any missing ingredients in the dish preparation.

[0046] As an example, an external knowledge base for AI models can be built based on professional knowledge, and information on how to prepare the target dish and the ingredients that are still missing in the preparation of the target dish can be generated through Retrieval-Augmented Generation (RAG).

[0047] RAG, or Retrieval and Augmentation, refers to a system where, when a model needs to generate text or answer questions, it first retrieves relevant information from a document collection (external knowledge base), then uses this retrieved information to guide text generation, thereby improving the quality and accuracy of predictions. It also enhances the interpretability and customizability of the generated results. RAG comprises three main processes: retrieval, augmentation, and generation. Retrieval involves obtaining relevant information from an external knowledge base based on the desired query content. Specifically, the query request is converted into a vector using an embedding model for comparison with relevant knowledge stored in a vector database. A similarity search identifies the top K most relevant matches to the query. Augmentation involves embedding the query content and retrieved relevant knowledge into a pre-defined prompt template. Generation involves inputting the augmented prompt content into the model to generate the desired output.

[0048] To improve the usability of the recommended recipes and encourage users to use them to prepare target dishes, embodiments of this application can also generate personalized recipes based on user needs. Specific examples will be provided below for explanation.

[0049] In one example, the server can obtain health tags representing the user's dietary preferences and use them as prompts for the AI ​​model. The AI ​​model then generates personalized information about how to prepare the target dish based on these health tags.

[0050] In this example, dietary preferences can be used to indicate a user's choices and aversions regarding ingredients and processing methods. Health labels can include at least: labels related to nutrition and health, such as low-fat and high-protein, sugar control, fat loss, muscle gain, etc.; and labels related to food preferences and dietary restrictions, such as halal, vegetarian, allergens, lactose intolerance, etc.

[0051] In practical applications, AI models can generate personalized recipes from different perspectives to adapt to the requirements of users' health labels, dynamically transforming the user's static health label into specific parameters in the cooking steps. Taking the health label "weight loss" as an example, the cooking method can be changed from "deep-frying" to "air frying," such as heating the air fryer to 180℃ for 20-25 minutes; the proportions of different ingredients in the recipe can also be adjusted, such as using a ratio of beef, potatoes, and onions of 1.2:1:1.5, adjusting the recipe to 120g beef, 100g potatoes, and 150g onions; the processing method of ingredients can also be adjusted, such as removing the skin from chicken legs; the amount of auxiliary ingredients in the recipe, especially the content of seasonings, can also be adjusted, such as halving the amount of white sugar or rock sugar; and so on, adjusting the recipe from multiple angles to meet the requirements of the user's health label.

[0052] In another example, if the knowledge information also includes knowledge about the substitution relationships of different ingredients, then when generating the preparation method information of the target dish, if it is determined that there is a substitute first ingredient in the target dish, and the existing ingredient information includes substitute ingredients for the first ingredient, then the substitute ingredients can be used to generate the preparation method information of the target dish.

[0053] In this example, the substitution relationships between different ingredients may be universal, applicable to most dishes when generating recipes based on substitution relationships; however, they may also be limited to specific dishes, where recipes can only be generated using ingredients with substitution relationships. This application does not specifically limit the association between ingredient substitution relationships and dishes. When it is determined that the ingredients required to prepare the target dish include a first ingredient with a substitution relationship, it can be further determined whether the substitution relationship of the first ingredient needs to be used for recipe generation. Specifically, if the existing ingredient information includes substitute ingredients with a substitution relationship to the first ingredient, the recipe can be adjusted based on the substitute ingredients to generate a personalized recipe that fully utilizes the user's existing ingredients and is sent to the user for recipe recommendation.

[0054] If the existing ingredient information includes alternative ingredients, the recipe can be adjusted directly. This can promote the consumption of existing ingredients, avoid burdening users with the decision of whether to substitute ingredients, and increase the likelihood that users will use the recommended recipe to make the target dish.

[0055] In another example, after providing the user with the preparation method information for the target dish, if the user submits correction information regarding the viewed preparation method information, the correction information can be used as a prompt for the AI ​​model. The AI ​​model will then adjust the preparation method information for the target dish based on the correction information and send the adjusted preparation method information back to the client for recipe recommendation until the recommended recipes can meet the user's dish preparation needs.

[0056] In this example, the correction information may include information related to user taste preferences and / or kitchenware usage preferences.

[0057] Taste preferences can be used to represent a user's taste and flavor experience, and may include at least: preferences related to the user's sense of taste, such as liking sweet and sour flavors, preferring spicy flavors, etc.; preferences related to the user's taste, such as having a mild taste, etc.; and preferences related to texture, such as liking crispy or preferring soft and chewy textures, etc.

[0058] For example, after viewing the preparation method information for a target dish through the client, a user can enter "I prefer a milder flavor." The client can then generate a modified flavor profile and send it to the server. This allows the AI ​​model to adjust the recipe, ensuring that the final preparation method provided to the user meets their preference for a milder flavor. This achieves truly personalized recipe recommendations tailored to the user's needs.

[0059] Kitchen appliance usage preferences can be used to indicate a user's kitchen appliance ownership and usage tendencies. For example, user A has an oven but no air fryer, user B has an air fryer, user C has both an oven and an air fryer but prefers to use the air fryer, and so on.

[0060] For example, for user A's health tag "weight loss," the generated recipe information might include cooking steps using an air fryer. If user A views this and then asks, "I don't have an air fryer, can you provide an oven-based recipe?", the client can generate corrected recipe information, specifying oven-based cooking, and send it to the server. The AI ​​model can then adjust the recipe, changing the air fryer steps to oven-based cooking and adjusting parameters like heating temperature and baking time. The adjusted recipe is then sent back to user A to meet their oven-based cooking needs. This recipe recommendation system, which incorporates user kitchen appliance preferences, ensures practicality, allowing users to prepare their target dishes using the recommended recipes, effectively pushing recipes and improving the user experience.

[0061] Understandably, the server can also save the correction information submitted by the user so that, in the subsequent recommendation process, personalized recipes can be generated based on the user's taste preferences, kitchenware preferences, and other information indicated by the correction information, so as to better meet the user's food preparation needs.

[0062] S203: Based on the missing ingredient information, the server determines recommended product information from the available products provided by the product information service isolation system.

[0063] S204: The server provides the client with information on how to prepare the target dish and recommended product information.

[0064] In this embodiment, while generating the preparation method information for the target dish, the AI ​​model can also determine missing ingredients to ascertain whether the existing ingredient information includes all the ingredients required to prepare the target dish. If all ingredients are included, the preparation method information can be sent to the client for display to the user; if only some ingredients are included, missing ingredient information can be generated, and product recommendations can be made accordingly. For example, in the above example, when generating preparation method information for the target dish "Curry Potato Beef," the ingredients required for preparation are determined to be potatoes, beef, curry cubes, and onions. The corresponding generated missing ingredient information includes curry cubes and onions.

[0065] After obtaining the missing ingredient information, the server can determine recommended product information from the available products provided by the product information service system, and make targeted product recommendations for the missing ingredients. For example, it can obtain recommended product information for curry cubes and recommended product information for onions based on at least one of the following factors: user's historical purchase habits, product reviews, and promotional activities.

[0066] In practical applications, the information users upload about their existing ingredients may not represent all the ingredients they currently have in stock. To improve the accuracy of recommendations for items in short supply, the server can provide the user with information on the missing ingredients for confirmation. Based on the ingredients the user selects from the list of missing ingredients, the server can then determine the recommended products. In other words, the server can recommend products based on the ingredients the user is truly missing. This more targeted approach helps increase the conversion rate of items in short supply.

[0067] It should be noted that if the user's cooking intention is provided to the user by the AI ​​model after intelligently identifying existing ingredient information as the main ingredient, the generated missing ingredient information will include the necessary auxiliary ingredients required to make the target dish. If the cooking intention is expressed by the user actively or under the guidance of a dialogue, the user usually has greater autonomy regarding the target dish they wish to make. The generated missing ingredient information may include both the main ingredient and auxiliary ingredients required to make the target dish, depending on the specific circumstances.

[0068] The client receives the preparation method information for the target dish and recommended product information from the server. It can simultaneously display both in the form of conversational content, providing an option to submit an online order for the recommended products. Alternatively, it can display both simultaneously on a webpage. The webpage includes a recipe display area showing the preparation method information for the target dish, and a product display area showing recommended product information and providing an option to add recommended products to the cart with one click. For example, the product display area might show product cards for curry cubes and onions, allowing users to select the desired items and submit an online order. The product information service system then fulfills the user's online order, delivering any missing items to complete the preparation of the target dish. This forms the complete service chain of this application, from uploading existing ingredients to generating the recipe and then ordering and delivering missing ingredients.

[0069] In summary, this application embodiment can intelligently analyze the user's uploaded ingredient information and expressed cooking intentions using an AI model to determine the target dish the user wishes to make. Furthermore, based on relevant knowledge about the target dish, it generates information on the preparation method and any missing ingredients. The preparation method information is provided to the user for recipe recommendation, and product recommendations are provided for the missing ingredients, ensuring a good match between recipe and product recommendations. Moreover, this application is based on consuming the user's existing ingredient reserves. By using existing ingredients, it drives the exposure and conversion of products containing missing ingredients. This product recommendation scheme is more targeted and better aligns with the daily consumption habits of most users, thus improving the user shopping experience.

[0070] It should be noted that the embodiments of this application may involve the use of user data. In practical applications, user-specific personal data may be used in the scheme described herein within the scope permitted by applicable laws and regulations, provided that it complies with the applicable laws and regulations of the country (e.g., with the user's explicit consent, with the user being properly notified, etc.).

[0071] Corresponding to the foregoing method embodiments, this application also provides a product recommendation device applied to a server, see [link to relevant documentation]. Figure 3 The device may include: The information receiving unit 301 is used to receive information on existing ingredients uploaded by the user and the cooking intention expressed. The information generation unit 302 is used to analyze the existing ingredient information and cooking intention through an artificial intelligence (AI) model, determine the target dish to be made, and generate the cooking method information of the target dish and the ingredient information that is still missing in making the target dish based on the knowledge information related to making the target dish. The recommended product determination unit 303 is used to determine recommended product information from the available products provided by the product information service system based on the missing ingredient information; The information providing unit 304 is used to provide the client with information on the preparation method of the target dish and the recommended product information.

[0072] Specifically, the information receiving unit can be used to receive existing ingredient information and cooking intention expressed in multimodal form in the form of images and / or text and / or voice. In this case, the AI ​​model includes a multimodal language model, or includes multiple single-modal language models adapted to the expression forms of the existing ingredient information and the cooking intention.

[0073] The existing ingredient information refers to images obtained by taking pictures of the ingredients that the user has stored, and the cooking intention is expressed in the form of text and / or voice.

[0074] Specifically, the information generation unit can be used to generate the preparation method information of the target dish in the following manner: obtaining a health label representing the user's dietary preferences as a prompt word for the AI ​​model, and the AI ​​model generating the preparation method information of the target dish based on the health label.

[0075] The knowledge information includes knowledge related to the substitution relationships of different ingredients. The information generation unit can be used to generate the preparation method information of the target dish in the following manner: based on the knowledge related to the substitution relationships, it is determined that there is a first ingredient that can be substituted in the target dish, and the existing ingredient information includes substitute ingredients for the first ingredient. The preparation method information of the target dish is generated using the substitute ingredients.

[0076] The device further includes: The correction information receiving unit is used to receive correction information submitted by the user regarding the preparation method information after providing the client with the preparation method information of the target dish. The correction information includes information related to the user's taste preferences and / or kitchenware usage preferences. The information generation unit can also be used to: use the correction information as a prompt word for the AI ​​model, and have the AI ​​model adjust the preparation method information of the target dish according to the correction information.

[0077] Specifically, the information receiving unit can be used to receive the cooking intention expressed by the user in the following manner: The existing ingredient information is used as the main ingredient in the dish, and the AI ​​model is used to identify the user's cooking intention. The identified cooking intention is provided to the user for confirmation, and the cooking intention confirmed by the user is taken as the cooking intention expressed by the user.

[0078] Specifically, if at least two cooking intentions are identified for the same main ingredient, the information receiving unit may further be used to: generate dish characteristic tags for the at least two cooking intentions and provide them to the user for confirmation of the cooking intentions; the dish characteristic tags are used to represent the characteristic information of the dish related to the cooking intentions.

[0079] Specifically, the recommended product determination unit can be used for: Provide the missing ingredient information to the user for confirmation; The recommended product information is determined based on the ingredients selected by the user from the missing ingredient information.

[0080] Corresponding to the foregoing method embodiments, this application also provides a product recommendation device applied to a client-side application. See [link to relevant documentation]. Figure 4 The device may include: The information sending unit 401 is used to receive existing ingredient information and the cooking intention expressed by the user, and send them to the server so that the server can analyze the existing ingredient information and cooking intention through an artificial intelligence (AI) model, determine the target dish to be made, and generate the cooking method information of the target dish and the missing ingredient information based on the knowledge information related to making the target dish; based on the missing ingredient information, it determines recommended product information from the available products provided by the product information service system. The information receiving unit 402 is used to receive and display the preparation method information of the target dish and the recommended product information sent by the server.

[0081] The device further includes: The session establishment unit is used to establish a session connection with the server in an intelligent dialogue manner and provide a session interface so as to obtain the existing ingredient information and the cooking intention through the session interface, and to display the preparation method information of the target dish and the recommended product information through the session interface.

[0082] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0083] And an electronic device, comprising: One or more processors; and A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the foregoing method embodiments.

[0084] A computer program product includes a computer program / computer executable instructions that, when executed by a processor in an electronic device, implement the steps of the method described in the foregoing method embodiments.

[0085] in, Figure 5An exemplary architecture of an electronic device is shown, which may include a processor 510, a video display adapter 511, a disk drive 512, an input / output interface 513, a network interface 514, and a memory 520. The processor 510, video display adapter 511, disk drive 512, input / output interface 513, network interface 514, and memory 520 can communicate with each other via a communication bus 530.

[0086] The processor 510 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in this application.

[0087] The memory 520 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 520 can store the operating system 521 for controlling the operation of the electronic device 500, and the basic input / output system (BIOS) 522 for controlling the low-level operations of the electronic device 500. Additionally, it can store a web browser 523, a data storage management system 524, and a product recommendation system 525, etc. The aforementioned product recommendation system 525 can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when implementing the technical solution provided in this application through software or firmware, the relevant program code is stored in the memory 520 and executed by the processor 510.

[0088] Input / output interface 513 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0089] Network interface 514 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0090] Bus 530 includes a pathway for transmitting information between various components of the device, such as processor 510, video display adapter 511, disk drive 512, input / output interface 513, network interface 514, and memory 520.

[0091] It should be noted that although the above-described device only shows the processor 510, video display adapter 511, disk drive 512, input / output interface 513, network interface 514, memory 520, bus 530, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.

[0092] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. 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 software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0093] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0094] The solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A product recommendation method, characterized in that, The method includes: Receive information about existing ingredients uploaded by users, as well as their cooking intentions; The existing ingredient information and cooking intention are analyzed by an artificial intelligence (AI) model to determine the target dish to be made. Based on the knowledge information related to making the target dish, the preparation method information of the target dish and the ingredient information that is still missing in making the target dish are generated. Based on the missing ingredient information, recommended product information is determined from the available products provided by the product information service system; Provide the client with information on how the target dish is prepared, as well as information on recommended products.

2. The method according to claim 1, characterized in that, The existing ingredient information and the cooking intention are expressed in a multimodal manner in the form of images and / or text and / or voice. The AI ​​model includes a multimodal language model, or includes multiple unimodal language models adapted to the expression of the existing ingredient information and the cooking intention.

3. The method according to claim 2, characterized in that, The existing ingredient information is obtained by taking pictures of the ingredients that the user has stored, and the cooking intention is expressed in the form of text and / or voice.

4. The method according to any one of claims 1 to 3, characterized in that, The information on the preparation method for generating the target dish includes: A health label representing the user's dietary preferences is obtained and used as a prompt word for the AI ​​model. The AI ​​model then generates information on the preparation method of the target dish based on the health label.

5. The method according to any one of claims 1 to 3, characterized in that, The knowledge information includes knowledge related to the substitution relationships of different ingredients, and the information on the preparation method for generating the target dish includes: Based on the knowledge related to the substitution relationship, it is determined that there is a first ingredient that can be substituted in the target dish, and the existing ingredient information includes substitute ingredients for the first ingredient. The preparation method information of the target dish is generated using the substitute ingredients.

6. The method according to any one of claims 1 to 3, characterized in that, After providing the client with the preparation method information of the target dish, the method further includes: Receive correction information submitted by the user regarding the preparation method information, the correction information including information related to the user's taste preferences and / or kitchenware usage preferences; The correction information is used as a prompt word for the AI ​​model, which then adjusts the preparation method information of the target dish based on the correction information.

7. The method according to any one of claims 1 to 3, characterized in that, Receive cooking intentions expressed by the user, including: The existing ingredient information is used as the main ingredient in the dish, and the AI ​​model is used to identify the user's cooking intention. The identified cooking intention is provided to the user for confirmation, and the cooking intention confirmed by the user is taken as the cooking intention expressed by the user.

8. The method according to claim 7, characterized in that, If at least two cooking intentions are identified for the same main ingredient, the method further includes: For the at least two cooking intentions, generate dish characteristic tags and provide them to the user for confirmation of the cooking intentions; the dish characteristic tags are used to represent the characteristic information of the dish related to the cooking intention.

9. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Provide the missing ingredient information to the user for confirmation; The recommended product information is determined based on the ingredients selected by the user from the missing ingredient information.

10. A product recommendation method, characterized in that, The method includes: The client receives information about existing ingredients uploaded by the user and their cooking intentions, and sends it to the server. The server then uses an artificial intelligence (AI) model to analyze the information and intentions to determine the target dish to be made. Based on knowledge related to making the target dish, the server generates information about the preparation method and any missing ingredients. Based on the missing ingredient information, the server determines recommended products from the available products provided by the product information service system. The system receives and displays the preparation method information of the target dish and the recommended product information sent by the server.

11. The method according to claim 10, characterized in that, The client establishes a session connection with the server via intelligent dialogue and provides a session interface to obtain information about the existing ingredients and the cooking intention, and to display information about the preparation method of the target dish and the recommended products through the session interface.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program performs the steps of the method described in any one of claims 1 to 11.

13. An electronic device, characterized in that, include: One or more processors; as well as A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method according to any one of claims 1 to 11.

14. A computer program product comprising a computer program / computer-executable instructions, characterized in that, When the computer program / computer-executable instructions are executed by a processor in an electronic device, they implement the steps of the method according to any one of claims 1 to 11.

Images