CN115221420A - Diet recommendation method and system based on user portrait - Google Patents
Diet recommendation method and system based on user portrait Download PDFInfo
- Publication number
- CN115221420A CN115221420A CN202210757287.7A CN202210757287A CN115221420A CN 115221420 A CN115221420 A CN 115221420A CN 202210757287 A CN202210757287 A CN 202210757287A CN 115221420 A CN115221420 A CN 115221420A
- Authority
- CN
- China
- Prior art keywords
- user
- information
- diet
- menu
- package
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a diet recommendation method and system based on a user portrait, which are characterized in that diet information of a user is obtained, diet characteristics of the user are obtained through extraction according to one or more of diet preference information, position information and age information of the user contained in the diet information of the user, the user portrait is generated based on the diet characteristics, and a diet menu and/or a package are recommended for the user according to the user portrait. The method provided by the embodiment analyzes the individual dietary preference, consumption characteristics and other information of the user, draws the user portrait, and recommends the menu or package for the user according to the drawn user portrait, so that the accuracy of recommending the menu and package is improved, the requirement of personalized recommendation of the user is met, the selection time of purchasing food materials, the menu or package by the user is reduced, and convenience is provided for the user to make or select the food.
Description
Technical Field
The invention relates to the field of smart city science and technology construction, in particular to a diet recommendation method and system based on user portrait.
Background
Along with the increasing abundance of life, people pay more and more attention to healthy diet nowadays, and not only pay more attention to food materials, but more and more people begin to try to make meals by themselves, so that the happiness of healthy self life is greatly improved.
With the appearance of offline shopping malls and a plurality of online shopping software, the problem of shopping and delivering primarily meets the requirements, but because the food materials are various and complete in variety, and the food materials correspond to menus and packages with different tastes and different functions, a user needs to spend a large amount of time on selecting food materials or selecting menus and packages every day.
If only the historical purchase information of the user is used as the recommended food material or recipe for the user, the accuracy of the recommended food material or recipe may be low due to uncertain factors such as the change of the commodity type and the change of the eating habits of the user, and the food material or recipe which is interested in the user cannot be quickly and accurately recommended for the user, so that the current catering related software cannot meet the personalized requirements of the user.
Therefore, the prior art is subject to further improvement.
Disclosure of Invention
In view of the defects in the prior art, the invention provides a diet recommendation method and system based on user portrait, and solves the problem that in the prior art, the dish buying software only has the dish buying function and lacks the function of providing exclusive food materials, recipes or package recommendations for the user, so that the personalized needs of the user cannot be met.
In a first aspect, the present embodiment discloses a diet recommendation method based on a user profile, including:
acquiring user diet information, and extracting diet characteristics from the user diet information; the user diet information includes: user dietary preference information, user location information, and age information;
generating a user representation containing a user dietary preference tag according to the dietary characteristics;
and recommending a menu and/or a package for the user according to the user portrait.
Optionally, before the step of obtaining the user diet information and extracting the user diet characteristics from the user diet information, the method includes:
collecting diet preference information and food material purchasing information of a plurality of users, and establishing a preference database; the food material purchasing information comprises address information of purchasing food materials and the purchasing quantity of each food material in a preset time period;
the step of generating a user representation containing tags of the user's dietary preferences based on the dietary characteristics comprises:
and matching the dietary characteristics with the dietary preference information and food material purchase information contained in the preference database to generate a user portrait corresponding to the dietary characteristics.
Optionally, before the step of generating a user representation including a tag of the user's dietary preference according to the dietary characteristics, the method further includes:
collecting address information of service stations for a plurality of users to purchase food, member IDs of users purchased by each service station and food stock information in each service station, and establishing a service station database;
collecting menu information uploaded by each user and establishing a menu database; the menu information includes: the recipe name, the recipe attribution cuisine of the recipe and the cooking step of the recipe;
collecting package information uploaded by each user, and establishing a package database; the package database includes: package ID, menu name and menu information contained in the package.
Optionally, the step of recommending a menu and/or a package for the user according to the user representation includes:
obtaining the evaluation information of each user on diet in the same time period of each service site of a service site database; the evaluation information includes: scoring the menu and the set of meals, browsing the menu and the set of meal quantity information, and evaluating the information of the menu and the food materials in the set of meals;
and combining the diet evaluation information of each user with the user portrait of each user to recommend a menu and/or a package to each user.
Optionally, the step of combining the diet evaluation information of each user with the user profile of each user to recommend the menu and/or package to the user includes:
and filtering the preference database information of the users by using a near-real-time collaborative filtering algorithm to generate interest vectors of a plurality of users in the same group for a plurality of recipes, querying a recipe database and a package database based on the interest vectors, outputting matching items, and outputting recommended packages and/or recipes according to weight sorting.
Optionally, the method further includes:
and updating the user portrait of the user at preset intervals according to the diet information of the user to obtain an updated user portrait, and recommending a menu and/or a package for the user according to the updated user portrait. The method further comprises the following steps:
and calculating the similarity between every two recipes by using a distributed algorithm, orderly arranging the similarity of the recipes to the weight ratio of the target type user to form a list, and pushing the corresponding or similar recipes to the user by combining the recipe information in the list contained by the recipe database.
In a second aspect, the present embodiment further provides a diet recommendation system based on a user profile, including:
the information extraction module is used for acquiring the diet information of the user and extracting diet characteristics from the diet information of the user; the user diet information includes: user dietary preference information, user location information, and age information;
the sketch generation module is used for generating a user sketch containing a user diet preference label according to the diet characteristics;
and the recommending module is used for recommending a menu and/or a package for the user according to the user portrait.
Optionally, the system further includes:
the preference database building module is used for collecting diet preference information and food material purchasing information of a plurality of users and building a preference database; the food material purchasing information comprises address information of purchasing food materials and the purchasing quantity of each food material in a preset time period;
and the portrait generation module generates a user portrait corresponding to the user diet information according to the diet preference information and the food material purchase information contained in the preference database.
Optionally, the recommending module includes: the system comprises a site information collection unit and a centralized recommendation unit;
the station information collection unit is used for acquiring the evaluation information of each user on diet in the same time period of each service station of the service station database; the evaluation information includes: scoring the menu and the set of meals, browsing the menu and the set of meal quantity information, and evaluating the information of the menu and the food materials in the set of meals;
and the centralized recommendation unit is used for combining the diet evaluation information of each user with the user portrait of each user to recommend a menu and a package to each user.
The invention provides a diet recommending method and system based on a user portrait, which are characterized in that diet information of a user is obtained, diet characteristics of the user are extracted according to one or more of diet preference information, position information and age information of the user contained in the diet information of the user, the user portrait is generated based on the diet characteristics, and a diet recipe and/or a package are recommended for the user according to the user portrait. The method provided by the embodiment analyzes the individual dietary preference, consumption characteristics and other information of the user, draws the user portrait, and recommends the menu or package for the user according to the drawn user portrait, so that the accuracy of menu and package recommendation is improved, the requirement of personalized recommendation of the user is met, the selection time of the user for purchasing food materials, menus or packages is reduced, and convenience is provided for the user in making or selecting the diet.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flowchart illustrating steps of a method for user profile based diet recommendation in accordance with an embodiment of the present invention;
FIG. 2 is a block diagram of a schematic structure of a user profile-based diet recommendation system according to an embodiment of the present invention;
fig. 3 is a schematic block diagram of an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
With the increasing popularization of scientific and technological products, various application programs are generated along with the needs of people, for example, at present, people can conveniently use various vegetable buying APPs or applets to realize the rapid ordering of various food materials, and then the food materials are delivered to the hands of users in the modes of takeaway or express delivery and the like, so that the vegetables can be bought without the users going out.
However, the current vegetable buying software only provides a vegetable buying function generally, and does not have a function of recommending a menu or a package for a user in a self-adaptive manner aiming at the user requirement, and the user needs to spend more time for selecting food materials of the required menu or food materials corresponding to the package each time, so that the efficiency is low, and the current software or network information cannot meet the requirements of learning and communication of the user.
In order to overcome the above problems, the present embodiment provides a diet recommendation method and system based on a user portrait, which implement recommending diets for users according to individual diet characteristics of the users by acquiring diet information of the users, generating user portraits of the users according to the diet information of the users, recommending recipes and packages for the users based on the user portraits, and pushing purchasing links of corresponding food materials for the users based on the recipes and packages selected by the users. Because big data analysis is utilized in the embodiment, a menu is recommended for the user based on the self preference of the user, and an exclusive package is recommended, so that the all-round requirements of the user are met in catering recommendation.
The embodiments of the present invention will be described in more detail with reference to the accompanying drawings.
The embodiment discloses a diet recommendation method based on user portrait, as shown in fig. 1, including:
s1, acquiring user diet information, and extracting diet characteristics from the user diet information; the user diet information includes: user dietary preference information, user location information, and age information.
In this step, when the user recommends diet by using the diet recommendation method provided by the present invention, the diet characteristics of the user may be first extracted according to the account information and the user basic information input by the user. Specifically, the account information input by the user may include: allergen information, geographical location information, age intervals, favorite dishes and other information, a menu and a package containing allergic food materials can be excluded according to the allergen information, and then the menu or the package in a corresponding area is recommended based on the geographical location information, for example: hot dry noodles can be recommended to be relevant in Wuhan areas, fried sauce noodles, roast ducks and the like in Beijing areas, and recipes such as pancake fruits, steamed stuffed buns and the like in Tianjin areas; depending on the age interval, different kinds of diets may be offered, such as: a student meal with rich nutrition is recommended for students, a healthy meal with low fat is recommended for the elderly, and the like, and if the user inputs a favorite diet, the meal can be recommended together.
Based on the content in the user diet information, diet characteristics can be extracted, and the diet characteristics can be classified according to different conditions, for example; according to the dietary characteristics of regions, the dietary characteristics of different regions are different (such as light, spicy or sweet), the dietary characteristics of different age groups (such as different dietary habits of young people and different dietary characteristics of children or old people), or the dietary characteristics of certain patients (such as diabetes patients and hypertension patients), so that the dietary characteristics can be acquired based on the individual dietary information.
And S2, generating a user portrait containing a user diet preference label according to the diet characteristics.
The user representation proposed in this step collects information related to the user, abstracts the concrete information of the user into tags according to the collected information, and concretes the user information by using the tags, thereby providing targeted services for the user. For example: when the purchase quantity of the peppers in the collected food material transaction data of the user exceeds the average purchase quantity of the user, the favorite taste of the user is described to be spicy, and Hunan dishes and Sichuan dishes are probably more suitable for being recommended to the user. If the purchase quantity of the peppers in the food material transaction quantity of the user is almost zero and the quantity of the fruits and the vegetables is large, the user may have a light diet preference at ordinary times, and the user is suitable for recommending a light menu or package to the user.
It is conceivable that the degree of matching between the user image drawn and the user himself/herself becomes higher as the dietary characteristics of the user extracted from the user dietary information increase, and that the degree of matching between the user image and the user himself/herself becomes lower as the amount of information decreases.
And S3, recommending a diet menu and/or package for the user according to the user portrait.
And recommending a corresponding diet menu or package for the user according to the user portrait drawn in the step. Specifically, the packages in the menu database and the package database are matched according to notes contained in the user portrait of each user, and the successfully matched menus and packages are recommended to the client.
The content of the menu can contain information such as pictures, characters, videos or audios, each menu can also have one or two of the pictures and the characters, the operation videos can be available or unavailable, each menu has a corresponding menu type, for example, belongs to meat dishes, vegetable dishes or Muslim menu, each menu also corresponds to an effect label to which the menu belongs, and the effect label comprises: losing weight, increasing muscle, maintaining beauty and keeping young and the like, and the menu is convenient to search under the corresponding effect label.
Further, in order to better render a user image for each user, the step of acquiring the user diet information and extracting the diet characteristics from the user diet information includes, before the step of acquiring the user diet information:
collecting diet preference information and food material purchasing information of a plurality of users, and establishing a preference database; the food material purchasing information comprises address information of purchasing food materials and the purchasing quantity of each food material in a preset time period; the users can be classified in the preference database according to the diet preferences of the users, and a plurality of user diet preference groups are obtained.
Further, in order to accurately draw the user portrait of the user, after the preference database is established, the step of generating the user portrait according to the user diet information comprises the following steps: and matching the dietary characteristics with the dietary preference information and food material purchase information contained in the preference database to generate a user portrait corresponding to the dietary characteristics.
After the user diet information is obtained, matching the user diet information with the labels of all the user diet preference groups in the preference database, matching the user preference group to which the user belongs, combining the diet information of the user with the labels corresponding to the user diet information in the user preference group, and giving the user portrait of the user.
Further, before the step of obtaining the diet information of the user and generating the user image based on the diet information, the method further includes:
collecting address information of service stations for a plurality of users to purchase food, member IDs of users purchased by each service station and food stock information in each service station, and establishing a service station database;
collecting menu information uploaded by each user and establishing a menu database; the menu information includes: the recipe name, the recipe attribution cuisine and the recipe cooking step;
collecting package information uploaded by each user and establishing a package database; the package database includes: package ID, recipe name and recipe information contained in the package.
Specifically, the preference database comprises a supplementary database of a member information base, and collects and stores preference data of members by using member IDs as unique identifiers, wherein member purchase information is collected and stored in the preference database; the member purchase information includes: the method comprises the steps of community service point address information, ID of specific dishes which do not want to be eaten, allergen names, ID of favorite dishes, number of favorite dishes, purchase times of favorite dishes and purchase order numbers.
In addition, the service station database contains: and the association database of the member information base comprises address information of the service station, all member IDs contained in the area and inventory information. The menu database contains menu information including menu ID, names of all dishes contained in the menu, menu details, menu affiliation cuisine and menu cooking steps. The package database contains: package information, menu information contained in a period set by a merchant, including package ID, all menu names, menu details and menu suggestion eating time period.
In order to better recommend recipes and packages according to the user figures, the method is also provided with a service station database, a recipe database and a package database, wherein the service station database can collect commodity purchasing data and commodity browsing information in the same service station and collect and analyze the information, so that users in the service station can be classified, eating habits of the same or similar new users are classified into one class, when the similar new users appear, the new users can be classified according to the purchasing commodity information and the commodity browsing information, and the recipes and packages are recommended according to the eating habits corresponding to the classified user types.
Various menus and various packages are respectively stored in the menu database and the package database, and the menus and the packages correspond to different user types, so that corresponding menu and package recommendation can be carried out on the user after the user portrait information of the user is obtained.
Specifically, the step of recommending a diet recipe and a package for the user according to the user portrait comprises the following steps:
obtaining the evaluation information of each user on diet in the same time period of each service site of a service site database; the evaluation information includes: scoring the menu and the set of meals, browsing the menu and the set of meal quantity information, and evaluating the information of the menu and the food materials in the set of meals;
and combining the diet evaluation information of each user with the user portrait of each user to recommend recipes and packages for each user.
In one embodiment, all member information in the service station database is obtained, and the behavior of each member on dishes/recipes/packages is recorded and scored. For example, the payment order is marked as 7, the browsed item is marked as 3, and the browsed item is marked as 5. For weight reduction calculations, behavior records and scores were kept for only the N most recent period (7 days). Based on the calculated information, the users are distributed to different partitions for user recommendation.
Further, the method can also combine the diet evaluation information of each user with the user portrait of each user to recommend the menu and the package to the user, and the method comprises the following steps:
and filtering the preference database information of the users by using a near-real-time collaborative filtering algorithm to generate interest vectors of a plurality of users in the same group for a plurality of recipes, querying a recipe database and a package database based on the interest vectors, outputting matching items, and outputting recommended packages and/or recipes according to weight sequencing.
In this embodiment, a near real-time Collaborative Filtering idea is used to recommend a package or a recipe for a user, and specifically, a main idea of Collaborative Filtering (CF) is to use other user groups having the same interest or historical behavior as the current user to predict what the current user may like or behaviors that may be generated by analyzing current preference or behavior information of the user groups. Therefore, according to the collected information in the service station database, generating interest vectors of a plurality of users in the same group for a plurality of recipes or packages, inquiring the recipe database and the package database based on the generated interest vectors, outputting matching items, and sorting and outputting recommended packages and/or recipes based on the matching items according to the weights.
Since the user may change in diet due to the user's own eating habits or based on changes in the occupied area, in order to maintain the accuracy of the recommendation, the method further includes:
and updating the user portrait of the user at preset time intervals according to the diet information of the user to obtain an updated user portrait, and recommending diet for the user according to the updated user portrait.
In the step, the historical user portrait of the user is updated based on the obtained latest user diet information to obtain an updated user portrait, and the user portrait changes along with the change of the user diet information, so that more accurate recommendation effect can be obtained by using the new user portrait to recommend the diet for the user.
In order to further realize accurate recommendation of the menu, the method further comprises the following steps:
and calculating the similarity between every two recipes by using a distributed method, orderly arranging the similarity of the recipes to the weight ratio of the target type user to form a list, and pushing the corresponding or similar recipes to the user by combining the recipe information contained in the list in the recipe database. Where the target category of users may be for a certain user group, such as: hyperglycemia population or student population.
Due to the fact that the types of the menus are various, different menus can be manufactured based on the same menu of different people, and particularly, corresponding menus and packages need to be manufactured specially for specific people, so that in the step, in order to better recommend the user, the similarity between the menus is compared, and more accurate menus are recommended for the user.
Further, after the user browses a plurality of recipes, purchase recommendation information for the user can be formed according to food materials corresponding to the plurality of browsed recipes. In the purchase recommendation information, the food materials are sorted according to the quantity contained in the menu browsed by the user, so that the user can quickly select and purchase favorite food materials. In order to provide different food material purchasing links for different users, the transaction processing module analyzes the menu information uploaded by each client, and performs food material purchasing recommendation on food materials contained in the uploaded menu information, so that purchasing recommendation information for the user is formed.
In order to facilitate a user to quickly purchase a suitable food material according to a favorite menu, a corresponding purchase link can be generated according to each food material listed in the menu, the user can automatically jump to a purchase interface of the food material by clicking the link, and the user can purchase the food material in the purchase interface.
In an implementation manner, the method disclosed in this embodiment may be implemented such that the user uploads information such as a menu and a package to a server equipped with a diet service platform, each user is equipped with a client, and the user logs in the service platform through the client to query recommended information and purchase food materials.
The package information uploaded to the server by the user includes a plurality of recipes, each package corresponds to one brief description, each brief description corresponds to one or more labels, and each label corresponds to the purpose and function of the package, for example, the label may be: the weight-losing set meal, the muscle-increasing set meal or the beauty set meal can be stored according to different categories of purposes and effects of the set meal. The simple description of the package may further include: contraindicated people of the set meal, whether food materials which are easy to be allergic exist, selection of cooking difficulty, data and description of nutritional ingredients, efficacy and the like, selection of recommended people (male and female, age group), remarks of the set meal and the like, wherein people with certain diseases can not eat the set meal or eat the set meal in a small amount. The package shorthand can also use the description of the recipe and select whether the package requires platform authentication for push (deducting certain points) and other information.
In order to facilitate the acquisition and processing of the information of each user, the client can also receive account editing information of each client user, update the user account information according to the account editing information, and authenticate and display the grade of the account according to the user account information.
When a client user logs in through a registered account, the client user can check historical browsing information in the account information, upload menu posts and package information, edit the uploaded menu posts and package information, and edit the account information, such as: setting information such as diet preference types, favorite cuisines and the like in account information, for example: and writing information such as preference for weight-losing packages and preference for Hunan dishes in the account information, and preferentially displaying the menu posts which accord with the cuisine or packages when searching the menu posts.
In one implementation mode, the diet service platform can further comprise a diet recommending plate, a community exchanging plate and a dish purchasing plate, a user logs in the service platform through an account, the user can search a menu and a package recommended by the service platform for the user in the diet recommending plate, the user can exchange cooking experience with other users in the community exchanging plate, the menu and the package made by the user can be published, and food materials can be purchased in the dish purchasing plate. The service platform comprises a user center unit, wherein information related to a user in the user center unit, such as posts for the user to publish a menu, menu packages and other information, the service platform correspondingly issues a point of credit after the user publishes the menu, packages or purchases food materials, and the user can exchange the credit. The user can also edit and process the post and menu package in the service platform.
And analyzing the menu information and the transaction data acquired from the big data collection unit, calculating the attention value of each menu of each different menu, sequencing the menus of the different menus according to the attention value, pushing the sequencing result to each client, establishing the corresponding relationship between the sequencing result and the food materials contained in the menu, and displaying the sequencing result of the menu corresponding to each food material on the purchase interface of each food material.
In order to better provide the exclusive recommendation service for the user, the information processing module collects the information of the user and the browsing information of the user, and pushes the collected information related to the user to the analysis processing unit. The information of the user can be information saved or filled in the account for the user, and the browsing information is menu, package or comment information browsed by the user within a preset time period.
In order to better recommend diet information to a user, a cuisine and food materials with high user attention are analyzed according to the collected big data, and information recommendation is carried out on the cuisine and food materials according to the cuisine and food materials with high attention, wherein the recommended information can be a favorite cuisine menu, a cuisine discussion post with high attention, a food material purchasing link with high attention and the like. The attention value can be calculated by weighting the information such as browsing times and purchasing times according to a preset proportion. And each user corresponds to an attention value, the attention value of each user to each menu and food material is combined with the user portrait of the user, and the menu and the food material are recommended to the user after the user portrait is corrected.
The service platform provided by the embodiment is based on the fact that the existing vegetable buying app software does not have related community communication edition blocks, provides a catering service platform with functions of vegetable buying, vegetable cooking, heart sharing, health guidance and one-stop service, and provides convenience for the life of a user.
On the basis of disclosing a user portrait-based diet recommendation method, the embodiment also provides a diet recommendation system based on a user portrait, as shown in fig. 2, the system includes:
the information extraction module 110 is configured to obtain diet information of a user, and extract diet features from the diet information of the user; the user diet information comprises: user dietary preference information, user location information, and age information;
a representation generation module 120 for generating a user representation containing a user dietary preference tag according to the dietary characteristics;
and the recommending module 130 is used for recommending a menu and/or a package for the user according to the user portrait.
Further, the system further comprises:
the preference database building module is used for collecting diet preference information and food material purchasing information of a plurality of users and building a preference database; the food material purchasing information comprises address information of purchasing food materials and the purchasing quantity of each food material in a preset time period;
and the portrait generation module generates a user portrait corresponding to the user diet information according to the diet preference information and the food material purchase information contained in the preference database.
Optionally, the recommending module includes: the system comprises a site information collection unit and a centralized recommendation unit;
the station information collection unit is used for acquiring the evaluation information of each user on diet in the same time period of each service station of the service station database; the evaluation information includes: scoring the menu and the set of food, browsing the menu and the set of food quantity information, and evaluating the information of the menu and the food in the set of food;
and the centralized recommendation unit is used for combining the diet evaluation information of each user with the user portrait of each user to recommend a menu and a package to each user.
Specifically, as shown in fig. 3, in an embodiment, the diet recommendation system provided in this embodiment is constructed by a server 11 and a client 12, where the client 12 is configured to collect relevant information of a user, and the server 12 is configured to process the collected relevant information.
The server 11 includes: an information extraction module 110, a representation generation module 120, and a recommendation module 130; the modules are respectively used for receiving menu information and diet information uploaded or input by each client user, extracting diet characteristics from the information related to the users, drawing a user portrait based on the diet characteristics of the users, and finally recommending diet for the users based on the user portrait.
A user can also set up (DIY) a self-time package menu according to self conditions, tastes and the like in the diet service platform, and in order to keep the freshness of dishes, the purchased food materials can be distributed in batches according to different food materials. The method and the system provided by the invention aim to fill the blank that the vegetable buying software in the market recommends to buy on the menu or the package at present, and provide the precise pushing of the dishes, the menu and the package aiming at the combination of the self diet information of the user and the big data. A complete dish set is scientific and nutritional, and is provided for a user, the user is not required to be busy selecting vegetables every day, only one set is needed, and batch distribution is realized in order to keep the freshness of dishes.
The invention provides a diet recommendation method and system based on a user portrait, which are characterized in that diet information of a user is obtained, diet characteristics of the user are obtained through extraction according to one or more of diet preference information, position information and age information of the user contained in the diet information of the user, the user portrait is generated based on the diet characteristics, and a diet menu and/or a package are recommended for the user according to the user portrait. The method provided by the embodiment analyzes the individual dietary preference, consumption characteristics and other information of the user, draws the user portrait, and recommends the menu or package for the user according to the drawn user portrait, so that the accuracy of menu and package recommendation is improved, the requirement of user personalized recommendation is met, the selection time for the user to purchase food materials, menus or packages is reduced, convenience is provided for the user to make or choose in the aspect of diet, and special packages are especially formulated for the special population for the user to select, such as weight-losing packages, body-building packages, muscle-building packages, old packages and the like. The set of meal is provided with a menu and dishes to be purchased.
It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is only limited by the appended claims
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A method for user profile-based dietary recommendation, comprising:
acquiring user diet information, and extracting diet characteristics from the user diet information; the user diet information includes: user dietary preference information, user location information, and age information;
generating a user portrait containing a user diet preference label according to the diet characteristics;
and recommending a menu and/or a package for the user according to the user portrait.
2. The user profile-based diet recommendation method of claim 1, wherein said step of obtaining user diet information and extracting the user's diet characteristics from said user diet information is preceded by the steps of:
collecting diet preference information and food material purchasing information of a plurality of users, and establishing a preference database; the food material purchasing information comprises address information of purchasing food materials and the purchasing quantity of each food material in a preset time period;
the step of generating a user representation containing tags of the user's dietary preferences based on the dietary characteristics comprises:
and matching the dietary characteristics with the dietary preference information and food material purchase information contained in the preference database to generate a user portrait corresponding to the dietary characteristics.
3. The user profile-based diet recommendation method of claim 2, wherein said step of generating a user profile including a user diet preference tag based on said diet characteristics is preceded by the steps of:
collecting address information of service stations for purchasing food materials by a plurality of users, member IDs of users purchased by each service station and food material inventory information in each service station, and establishing a service station database;
collecting menu information uploaded by each user and establishing a menu database; the menu information comprises: the recipe name, the recipe attribution cuisine and the recipe cooking step;
collecting package information uploaded by each user and establishing a package database; the package database includes: package ID, menu name and menu information contained in the package.
4. The user profile-based meal recommendation method of claim 3, wherein the step of recommending a recipe and/or package for a user based on the user profile comprises:
obtaining the evaluation information of each user on diet in the same time period of each service site of a service site database; the evaluation information includes: scoring the menu and the set of meals, browsing the menu and the set of meal quantity information, and evaluating the information of the menu and the food materials in the set of meals;
and combining the diet evaluation information of each user with the user portrait of each user to recommend a menu and/or a package for each user.
5. The user profile-based diet recommendation method of claim 4, wherein the step of combining diet evaluation information of each user and the user profile of each user to recommend a menu and/or package to the user comprises:
and filtering the preference database information of the users by using a near-real-time collaborative filtering algorithm to generate interest vectors of a plurality of users in the same group for a plurality of recipes, querying a recipe database and a package database based on the interest vectors, outputting matching items, and outputting recommended packages and/or recipes according to weight sequencing.
6. The user profile-based meal recommendation method of claim 1, further comprising:
and updating the user portrait of the user at preset intervals according to the diet information of the user to obtain an updated user portrait, and recommending a menu and/or a package for the user according to the updated user portrait.
7. The user profile-based meal recommendation method of claim 2, further comprising:
and calculating the similarity between every two recipes by using a distributed algorithm, orderly arranging the similarity of the recipes to the weight ratio of the target type user to form a list, and pushing the corresponding or similar recipes to the user by combining the recipe information in the list contained by the recipe database.
8. A user profile-based diet recommendation system, comprising:
the information extraction module is used for acquiring the diet information of the user and extracting diet characteristics from the diet information of the user; the user diet information includes: user diet preference information, user location information, and age information;
the portrait generation module is used for generating a user portrait containing a user diet preference label according to the diet characteristics;
and the recommending module is used for recommending a menu and/or a package for the user according to the user portrait.
9. The user representation-based diet recommendation system of claim 8, wherein said system further comprises:
the preference database building module is used for collecting diet preference information and food material purchasing information of a plurality of users and building a preference database; the food material purchasing information comprises address information of purchasing food materials and the purchasing quantity of each food material in a preset time period;
and the portrait generation module generates a user portrait corresponding to the user diet information according to the diet preference information and the food material purchase information contained in the preference database.
10. A user profile based diet recommendation system according to claim 8 or 9, wherein said recommendation module comprises: the system comprises a site information collection unit and a centralized recommendation unit;
the station information collection unit is used for acquiring the evaluation information of each user on diet in the same time period of each service station of the service station database; the evaluation information includes: scoring the menu and the set of meals, browsing the menu and the set of meal quantity information, and evaluating the information of the menu and the food materials in the set of meals;
and the centralized recommendation unit is used for combining the diet evaluation information of each user with the user portrait of each user to recommend recipes and packages to each user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210757287.7A CN115221420A (en) | 2022-06-30 | 2022-06-30 | Diet recommendation method and system based on user portrait |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210757287.7A CN115221420A (en) | 2022-06-30 | 2022-06-30 | Diet recommendation method and system based on user portrait |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115221420A true CN115221420A (en) | 2022-10-21 |
Family
ID=83609430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210757287.7A Pending CN115221420A (en) | 2022-06-30 | 2022-06-30 | Diet recommendation method and system based on user portrait |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115221420A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116628040A (en) * | 2023-05-16 | 2023-08-22 | 广东鸿智智能科技股份有限公司 | Big data-based cooking menu acquisition and updating method |
| CN117670439A (en) * | 2023-12-07 | 2024-03-08 | 深圳数拓科技有限公司 | Restaurant recommendation method and system based on user portrait |
-
2022
- 2022-06-30 CN CN202210757287.7A patent/CN115221420A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116628040A (en) * | 2023-05-16 | 2023-08-22 | 广东鸿智智能科技股份有限公司 | Big data-based cooking menu acquisition and updating method |
| CN116628040B (en) * | 2023-05-16 | 2024-05-07 | 广东鸿智智能科技股份有限公司 | Big data-based cooking menu acquisition and updating method |
| CN117670439A (en) * | 2023-12-07 | 2024-03-08 | 深圳数拓科技有限公司 | Restaurant recommendation method and system based on user portrait |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104994747B (en) | For providing the system and method for flavor suggestion and enhancing | |
| US7373318B2 (en) | Information recommendation apparatus and information recommendation system | |
| US20170316488A1 (en) | Systems and Methods of Food Management | |
| CN115221420A (en) | Diet recommendation method and system based on user portrait | |
| KR102371787B1 (en) | System for providing customized dietary management service | |
| CN103858142A (en) | Store information provision system | |
| JP2002041670A (en) | Device and system for recommending information | |
| CN110020186A (en) | A kind of dining room recommended method and system | |
| JP6413508B2 (en) | Information recommendation program and information processing apparatus | |
| CN111899047A (en) | Resource recommendation method and device, computer equipment and computer-readable storage medium | |
| US20230215293A1 (en) | System and method for designing food and beverage flavor experiences | |
| CN112597394A (en) | Menu recommendation method and device, storage medium and electronic device | |
| JP7096056B2 (en) | Shopping support system, shopping support server, program and user terminal. | |
| CN116645141B (en) | Multi-dimensional feature-fused chain store site selection recommendation method | |
| CN109741125A (en) | Recommend method and device, the storage medium, electronic device of vegetable | |
| CN112951374B (en) | Data analysis method, system, computer device and storage medium | |
| CN110874785A (en) | Method, device and equipment for determining meal package information | |
| CN111597434B (en) | Takeout recommendation method, system, device and medium based on user portrait | |
| KR102200391B1 (en) | Food recommendation service system based on analysis of user eating habits and method thereof | |
| KR101926579B1 (en) | Method for providing food favorite finding service with comparing taste from different company | |
| CN107392527B (en) | Article distribution method and article distribution scheme determination device | |
| JP2020197888A (en) | Eating-and-drinking supporting system | |
| KR20220014499A (en) | A taste network generation system based on user experience using CNN learning | |
| CN111598737A (en) | Method and system for automatically recommending dishes for customers | |
| CN112232916B (en) | Commodity recommendation method, device and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |