CN117670439A

CN117670439A - Restaurant recommendation method and system based on user portrait

Info

Publication number: CN117670439A
Application number: CN202311666487.2A
Authority: CN
Inventors: 郑山桥; 李凯
Original assignee: Shenzhen Shutuo Technology Co ltd
Current assignee: Shenzhen Shutuo Technology Co ltd
Priority date: 2023-12-07
Filing date: 2023-12-07
Publication date: 2024-03-08

Abstract

The invention discloses a restaurant recommendation method and a system based on user portraits, comprising the following steps: acquiring user historical order information, and performing attribute analysis according to the user historical order information to obtain attribute analysis result information; performing user preference analysis according to the user history order information and the attribute analysis result information, and constructing a user preference portrait; acquiring user history browsing information, analyzing user interests according to the user history browsing information, and constructing user interest portraits; predicting the next interest point of the target user according to the user interest portrait and the user preference portrait to obtain interest point prediction information; and acquiring real-time weather information, and carrying out catering recommendation by combining the user preference portrait, the user interest portrait and the interest point prediction information. The diversified demands of the users are comprehensively and accurately understood and reflected by the user portrait-based method, and the quality and individuation level of the catering recommended service are improved.

Description

Restaurant recommendation method and system based on user portrait

Technical Field

The invention relates to the technical field of catering recommendation, in particular to a catering recommendation method and system based on user portraits.

Background

With the explosive growth of social media and mobile applications, people's expectations for dining selection are gradually rising. However, conventional catering recommendation systems rely mainly on the past behavior data of users, and such methods have certain limitations. The tastes, interests and social backgrounds of users are multi-dimensional and dynamically changing, and it is difficult for conventional methods to fully grasp these personalized features.

The catering recommendation system in the current market often faces the problems of information overload and insufficient individuation. The user may feel trouble with excessive choices, and conventional recommendation systems often cannot accurately grasp the diversified demands of the user, so that the recommended dining choices do not always meet the actual expectations of the user.

Therefore, there is an urgent need for a more intelligent, personalized food recommendation method to better meet the user's diverse taste preferences, social needs, and real-time environmental changes.

Disclosure of Invention

The invention overcomes the defects of the prior art, and provides a restaurant recommendation method and a restaurant recommendation system based on user portraits, which aim at improving the accuracy and individuation degree of restaurant recommendation.

In order to achieve the above object, the first aspect of the present invention provides a restaurant recommendation method based on user portraits, including:

Acquiring user historical order information, and performing attribute analysis according to the user historical order information to obtain attribute analysis result information;

performing user preference analysis according to the user history order information and the attribute analysis result information, and constructing a user preference portrait;

acquiring user history browsing information, analyzing user interests according to the user history browsing information, and constructing user interest portraits;

predicting the next interest point of the target user according to the user interest portrait and the user preference portrait to obtain interest point prediction information;

and acquiring real-time weather information, and carrying out catering recommendation by combining the user preference portrait, the user interest portrait and the interest point prediction information.

In this solution, the attribute analysis is performed according to the user history order information to obtain attribute analysis result information, which specifically includes:

acquiring user history order information, extracting characteristics of the user history order, and extracting types, frequencies, prices and time characteristics of food in the user history order to obtain history order characteristic information;

presetting a cuisine category, carrying out cuisine attribute analysis according to the historical order characteristic information, calculating Euclidean distance between each order and the preset cuisine category, and judging with a preset threshold value to obtain cuisine attribute analysis information;

Acquiring order feedback information according to the historical order information, constructing a semantic analysis model based on a natural language processing technology, inputting the order feedback information into the semantic analysis model for feedback information semantic analysis, and obtaining semantic analysis result information;

performing taste attribute pre-analysis by combining the historical order feature information and the vegetable attribute analysis information, extracting the taste features of dishes in each order and the taste features of each vegetable, and counting the number of the taste features of the dishes in the order to obtain pre-analysis result information;

performing taste attribute analysis according to the semantic analysis result information and the pre-analysis result information to obtain taste attribute analysis information;

carrying out user dining time attribute analysis according to the historical order feature information, and dividing each order according to time to obtain time attribute analysis information;

according to the historical order characteristic information of the user, carrying out order price attribute analysis, analyzing the total price of each order and the price of the food, and counting the frequency of each price to obtain price attribute analysis information;

and combining the cuisine attribute analysis information, the taste attribute analysis information, the time attribute analysis information and the price attribute analysis information to obtain attribute analysis result information.

In this solution, the user preference analysis is performed according to the user history order information and the attribute analysis result information, and a user preference portrait is constructed, specifically:

acquiring historical order information and attribute analysis result information of a user;

obtaining the taste category and the menu category of the target user according to the attribute analysis result information, and analyzing by combining the user history order information based on a principal component analysis method to obtain principal component analysis information;

sorting all taste categories and all menu categories according to the main component information, and carrying out style preference analysis according to the sorting result to obtain style preference analysis information;

obtaining time attribute analysis information according to the attribute analysis result information, and carrying out time preference analysis by combining price attribute analysis information based on a statistical algorithm;

counting the dining frequency of the users in each time period, and performing time preference analysis by taking the dining frequency as a time preference analysis index to obtain time preference analysis information;

obtaining price attribute analysis information according to the attribute analysis result information, and drawing a total price trend chart and a unit price trend chart according to the price attribute analysis information;

according to the total price trend graph and the monovalent trend graph, price preference analysis is carried out on the target client, and the total price preference and the monovalent preference of the target client are analyzed to obtain price preference analysis information;

Combining the style preference analysis information, the time preference analysis information and the price preference analysis information to form user preference analysis information, and carrying out user preference grade evaluation;

presetting a preference level evaluation index, analyzing the user preference analysis information and the preference level evaluation index, evaluating the user preference style, the dining time and the price dimension, and endowing the user preference style, the dining time and the price dimension with corresponding preference levels to obtain preference level evaluation information;

and constructing a user preference portrait by combining the preference level evaluation information and the user preference analysis information.

In this scheme, the user interest analysis is performed according to the user history browsing information, and a user interest portrait is constructed, specifically:

acquiring historical browsing information of a user, and extracting characteristics of the historical browsing information of the user, wherein the characteristics comprise browsed food, browsing frequency and browsing time period, so as to obtain the historical browsing characteristic information;

constructing an associated interest analysis model according to an association rule mining algorithm, and inputting the historical browsing characteristic information into the associated interest analysis model for analysis to obtain associated interest analysis information;

setting interest labels according to the associated interest analysis information, wherein the interest labels are divided into time interest labels and meal interest labels, and setting corresponding interest weights;

Carrying out user interest preference degree analysis according to the interest labels and the interest weights and combining the associated interest analysis information and the historical browsing characteristics to obtain preference degree analysis information;

based on a clustering algorithm, carrying out user preference feature analysis according to the associated interest analysis information and the preference degree analysis information, and analyzing preference characteristics and types of target users to obtain preference feature analysis information;

and constructing the user interest portrait according to the preference degree analysis information, the associated interest score information and the preference characteristic analysis information.

In the scheme, the next interest point prediction is performed on the target user according to the user interest portrait and the user preference portrait, specifically:

the preset attribute category comprises a place attribute and a date attribute, wherein the place attribute is divided into a residence, a leisure and an office, and the date attribute is divided into a holiday and a workday;

acquiring historical dining position information and dining time information of a user, and carrying out attribute analysis by combining preset attribute categories to obtain dining attribute analysis information;

presetting time intervals, constructing a user access sequence comprising time, position and meal through meal attribute analysis information, and analyzing the behavior rule of a target user to obtain behavior rule analysis information;

Based on NGCF, constructing a long-short-term preference analysis model, acquiring user interest portraits and user preference portraits, and inputting dining attribute analysis information and behavior rule analysis information into a long-short-term preference score model to perform long-short-term preference analysis on a target user to acquire long-short-term preference analysis information;

constructing an interest point prediction model, and inputting the long-short-period preference analysis information and the behavior rule analysis information as initial information into the interest point prediction model to predict the next interest point;

calculating the attention score of each interest point according to the long-short-term preference analysis information and the behavior rule analysis information through an attention mechanism, taking the attention score as an interest point weight, and forming an interest decision vector through weighted fusion;

carrying out multi-layer nonlinear transformation on the interest decision vector and the interest points based on the multi-layer perceptron, and calculating recommendation scores of the interest points to obtain candidate interest point prediction information;

sorting according to the recommended scores of the candidate interest points, presetting a selection threshold, and selecting a final prediction result through the selection threshold to obtain the interest point prediction information.

In the scheme, the method for acquiring the real-time weather information and carrying out catering recommendation by combining the user preference portrait, the user interest portrait and the interest point prediction information comprises the following specific steps:

Acquiring real-time weather information, user preference portraits, user interest portraits and interest point prediction information;

the method comprises the steps of obtaining preference images and interest images of various different crowds based on big data retrieval to form a comparison data set, classifying the comparison data set, calculating Manhattan distances of the various crowds, and judging the Manhattan distances with a preset threshold value to obtain crowd class classification information;

integrating preferences and interests of various crowds, fusing similar interests and preferences, combining time characteristics to form a feature sequence, and constructing a feature map;

performing similarity calculation on the user preference portrait, the user interest portrait and the feature map, judging with a preset threshold, and analyzing similar crowds of the target user to obtain similar crowd information;

extracting the sequence of the feature map according to the similar crowd information, and extracting the feature sequence of the similar crowd to obtain similar feature sequence information;

performing proper catering analysis according to the real-time weather information, and analyzing proper catering under the current weather characteristics to obtain proper catering analysis information;

constructing a catering recommendation model, and recommending catering according to similar feature sequence information, suitable catering analysis information, user preference portraits, user interest portraits and interest point prediction information to obtain catering recommendation information;

And recommending the catering to the target user according to the catering recommendation information, wherein the catering recommendation comprises places, time and catering types.

In a second aspect, the present invention provides a user portrait-based food and beverage recommendation system, which includes: the food and beverage recommendation method based on the user portrait comprises a memory and a processor, wherein the memory contains a food and beverage recommendation method program based on the user portrait, and when the food and beverage recommendation method program based on the user portrait is executed by the processor, the following steps are realized:

Drawings

In order to more clearly illustrate the technical solutions of embodiments or examples of the present invention, the drawings that are required to be used in the embodiments or examples of the present invention will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the drawings without inventive efforts for those skilled in the art.

FIG. 1 is a flowchart of a restaurant recommendation method based on user portraits according to an embodiment of the present invention;

FIG. 2 is a flowchart of a method for recommending food and beverage based on user portraits according to an embodiment of the present invention;

FIG. 3 is a block diagram of a food recommendation system based on user portraits according to an embodiment of the present invention;

the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.

as shown in FIG. 1, the invention provides a restaurant recommendation method flow chart based on user portraits, which comprises the following steps:

s102, acquiring user history order information, and performing attribute analysis according to the user history order information to obtain attribute analysis result information;

It should be noted that, first, by obtaining the historical order information of the user, the dining preference and the behavior pattern of the user are obtained, including the key data such as the type, frequency, price and dining time of the ordered dishes. And secondly, carrying out detailed analysis on the historical orders through feature extraction, extracting various features including the types, the frequencies, the prices, the time and the like of the meal, establishing multidimensional features of user behaviors, and providing a basis for subsequent analysis. And then, classifying the order by presetting the menu category, and judging which menu the order belongs to, so that menu attribute analysis information is obtained, the taste preference of a user is refined, and the accurate cognition of the menu is improved. Further, by acquiring order feedback information and constructing a semantic analysis model by using a natural language processing technology, deep understanding of user feedback is realized, subjective evaluation and preference of a user on dishes are mined, and the semantic level of user portraits is enriched. And then, by combining the historical order information and the vegetable attributes, the taste characteristics are extracted, the number of the taste characteristics is counted, a basis is provided for subsequent taste attribute analysis, the taste preference of the user can be deeply displayed, and more personalized taste reference is provided for catering recommendation. Meanwhile, the attribute analysis of the dining time of the user is performed, orders are divided according to time, so that the habit of the dining time of the user in one day is known in depth, the preference of the dining time of the user is understood, and the consideration of time is provided for reasonable recommendation. Finally, the attribute analysis of the price of the order is carried out, the total price of the order and the price of the food are analyzed, the frequency of each price is counted, the acceptance degree of the user to the intervals with different prices is obtained, and the price guidance is provided for recommendation.

S104, carrying out user preference analysis according to the user history order information and the attribute analysis result information, and constructing a user preference portrait;

Firstly, determining the taste category and the dish category of a target user according to attribute analysis result information, and analyzing the historical order information of the user by adopting a principal component analysis method to obtain principal component analysis information. And then, sorting the taste categories and the menu categories according to the principal component analysis information, and carrying out style preference analysis on the sorting result to obtain the preference degree of the user on the taste and the menu, so as to accurately understand the taste tendency of the user and the preference of different menu. And then, obtaining time attribute analysis information according to the attribute analysis result information, and carrying out time preference analysis by combining a statistical algorithm and the price attribute analysis information. And the dining time tendency of the user is known by counting the dining frequency of the user in each time period. Meanwhile, a total price trend chart and a unit price trend chart are drawn by utilizing price attribute analysis information, and a reference is provided for subsequent price preference analysis. In price preference analysis, target clients are analyzed according to the total price trend graph and the unit price trend graph so as to know the preference degree of the user on the total price and the unit price, and the budget, the consumption habit and the hobbies of the user can be better understood. And then, combining style preference analysis information, time preference analysis information and price preference analysis information to form comprehensive user preference analysis information, and performing user preference grade evaluation. Through preset preference level evaluation indexes, the corresponding preference levels of the users in the style, the dining time and the price dimension are given, and detailed preference level evaluation information is obtained. Finally, by combining the preference level evaluation information and the user preference analysis information, a preference portrait of the user is constructed so as to better meet the personalized requirements of the user.

S106, acquiring user history browsing information, analyzing user interests according to the user history browsing information, and constructing user interest portraits;

It should be noted that, first, feature extraction is performed on the historical browsing information of the user, including key information such as browsed meal, browsing frequency and browsing time period, so as to obtain the historical browsing feature information. Then, an association interest analysis model is constructed by using an association rule mining algorithm, historical browsing characteristic information is input into the model for analysis, association interest analysis information is obtained, potential association rules in the historical browsing behaviors of the user are mined, and therefore interest association of the user is understood. According to the related interest analysis information, setting interest labels, dividing the interest labels into time interest labels and meal interest labels, and setting corresponding interest weights, so that the interests of users in time and meal are expressed more accurately. And then, analyzing the preference degree of the user interests by utilizing the interest tags and the interest weights in combination with the associated interest analysis information and the historical browsing characteristics, obtaining the preference degree of the user on different interest tags, and quantifying the preference of the user on the interests of time and dinner. And based on a clustering algorithm, carrying out user preference characteristic analysis according to the associated interest analysis information and the preference degree analysis information. Through clustering, preference characteristics and types of users are identified. Finally, the system constructs interest portraits of the users according to the preference degree analysis information, the associated interest analysis information and the preference feature analysis information, integrates the interests of the users in time and food aspects, and provides more comprehensive user understanding so as to better realize personalized recommendation service.

S108, predicting the next interest point of the target user according to the user interest portrait and the user preference portrait to obtain interest point prediction information;

It should be noted that, first, the system presets attribute categories including a place attribute and a date attribute, wherein the place attribute is classified into a house, a leisure and an office, and the date attribute is classified into a holiday and a workday. And then, acquiring historical dining position information and dining time information of the user, and carrying out attribute analysis by combining preset attribute categories to obtain dining attribute analysis information, so as to refine dining habits of the user at different places and dates and provide a basis for subsequent behavior rule analysis. And constructing an access sequence of the user through a preset time interval, wherein the access sequence comprises information such as time, position, meal and the like. Through analyzing the behavior rules of the target user, the dining habits of the user at different time periods and places are known in depth. Next, a long-short term preference analysis model is built based on the NGCF, interest images and user preference images of the user are obtained, long-term and short-term interest preferences of the user are understood, and accurate prediction of the user interests is improved. And further constructing an interest point prediction model, and inputting the long-short-period preference analysis information and the behavior rule analysis information into the model as initial information to predict the next interest point. And calculating the attention score of each interest point through an attention mechanism, taking the attention score as an interest point weight, and forming an interest decision vector through weighted fusion. And carrying out multi-layer nonlinear transformation on the interest decision vector and the interest points based on the multi-layer perceptron, and calculating the recommendation score of each interest point to obtain the prediction information of the candidate interest points. Finally, through the mode of sorting and setting a selection threshold value, the next interest point prediction is finally carried out on the target user so as to be used for recommending the catering of interest of the user, and catering recommendation which meets the personalized requirements of the user is provided for the user.

It should be noted that, the interest point prediction model is a frame constructed according to the LSTM, and on the basis of the frame, an attention mechanism and a multi-layer perceptron are added to perform model optimization, so as to improve the prediction capability of the model and obtain an accurate interest point prediction result.

Further, acquiring real-time position information of a user, extracting a diet high-frequency region of the target user through a user interest portrait of the target user, and judging with the real-time position information of the user to obtain user position judgment information; if the user position judgment information is not in the diet high-frequency area, calculating the real-time position information of the user and the space distance of the diet high-frequency area, judging with a threshold value, judging whether the user position judgment information is in a new diet area, and obtaining diet area judgment information; carrying out user state analysis according to the diet area judgment information and the user real-time position information to obtain user real-time state analysis information; judging whether regional special catering recommendation is carried out according to the real-time state analysis information of the user, and obtaining special catering recommendation judgment information; if the special catering recommendation judging information is recommendation, regional special catering information of the current position is obtained according to the real-time position information of the user, the recommendation is carried out according to the regional special catering information, and dining experience of the user is improved.

It is to be noted that, by analyzing the space distance between the real-time position of the user and the diet high-frequency area, judging whether the user is out of the diet high-frequency area, analyzing the user state, judging whether the user is in a travel state or a business trip state, and the like, carrying out regional characteristic diet recommendation according to the state judgment result, integrating regional culture factors, analyzing the characteristic diet culture of the area where the user is located, and providing the user with diet recommendation more conforming to local taste and culture atmosphere.

S110, acquiring real-time weather information, and carrying out catering recommendation by combining the user preference portrait, the user interest portrait and the interest point prediction information;

Firstly, a comparison data set is formed by the preference portraits and the interest portraits of various different crowds through big data retrieval, and classification is carried out. And calculating Manhattan distances among various crowds and judging with a preset threshold value to obtain crowd category division information. And then, integrating preferences and interests of various crowds, fusing similar interests and preferences, combining time characteristics to form a characteristic sequence, constructing a characteristic map, and understanding the similarity among different crowds. And then, carrying out similarity calculation on the user preference portrait, the user interest portrait and the feature map, judging with a preset threshold, analyzing the similar crowd of the target user to obtain similar crowd information, carrying out sequence extraction on the feature map through the similar crowd information, extracting the feature sequence of the similar crowd, and mapping the behavior rule and the eating habit of the similar crowd, thereby providing reference for personalized catering recommendation. And then, analyzing the proper dining under the current weather characteristics according to the real-time weather information, so as to provide dining recommendation which is more close to the current demands of users according to weather conditions, for example, proper eating of hot pot in cold days, proper eating of light food in hot days and the like. Finally, constructing a food and beverage recommendation model, and recommending food and beverage according to similar characteristic sequence information, suitable food and beverage analysis information, user preference portraits, user interest portraits and interest point prediction information to obtain food and beverage recommendation information, and recommending food and beverage to target users, wherein the food and beverage recommendation information comprises places, time and food and beverage types so as to meet personalized food and beverage requirements of the users. Through various considerations, the catering recommendation service which meets the actual requirements better is provided for users.

Further, obtaining similar crowd information, and obtaining historical purchase catering information of similar crowd according to the similar crowd information; performing association analysis of the food according to the historical purchase food information, and calculating attention scores among the food and beverage based on an attention mechanism, wherein the attention scores are used as confidence of the association analysis to obtain association analysis information; extracting the occurrence frequency of the associated catering according to the association analysis information, judging by a preset threshold, and defining the associated catering which is larger than the preset threshold as a high-associated catering to obtain high-associated catering analysis information; extracting features of the historical purchase catering information to obtain historical purchase catering feature information; presetting a plurality of attribute tags, and carrying out association classification on historical purchased food and the corresponding attribute tags based on a clustering algorithm to obtain food attribute classification information; constructing a restaurant knowledge graph according to the restaurant purchase time by combining the historical purchase restaurant information, the high-correlation restaurant analysis information and the dining attribute classification information; acquiring user real-time behavior information, including user real-time purchase information, user real-time access information and user real-time interaction information; extracting features of the user real-time purchasing behavior information, extracting user real-time purchasing dining information and browsing dining information, and obtaining user real-time behavior feature information; calculating the similarity between the real-time behavior characteristic information of the user and the restaurant knowledge graph, judging the similarity with a preset threshold value, and acquiring similar restaurants according to a judging result to obtain similar restaurant information; searching according to the similar food and beverage information and combining with a food and beverage knowledge graph, extracting food and beverage after a food and beverage path where the similar food and beverage is located as potential interest food and beverage, and obtaining potential interest food and beverage information; comparing and judging according to the high-correlation catering analysis information and the potential interest catering information, and judging whether high-correlation catering exists or not; if the high-correlation catering exists, recommending the corresponding high-correlation catering; if the high-correlation catering does not exist, extracting the occurrence frequency of the potential interest catering in a catering knowledge graph, taking the occurrence frequency as a recommendation weight, and carrying out weighted calculation on the potential interest catering information; and carrying out food recommendation decision according to the weighted calculation result, and recommending potential intention food to the user, so that the user preference is closed, and the user experience is improved.

as shown in fig. 2, the present invention provides a restaurant recommendation flowchart of a user portrait-based restaurant recommendation method, which includes:

s202, analyzing the attribute of the target user, analyzing the style preference, the time preference and the price preference of the target user according to the attribute analysis result, and constructing a preference portrait;

s204, carrying out user interest analysis on the target user and constructing interest portraits;

s206, carrying out next interest point prediction by combining the preference portrait and the interest portrait;

s208, analyzing characteristics of target users, and analyzing catering possibly interested by the users through similar crowds;

s210, analyzing catering suitable for the current meteorological environment by combining real-time meteorological;

s212, recommending the catering, including places, time and catering types.

Further, presetting an information acquisition time interval, and acquiring restaurant order information of a target user according to the information acquisition time interval; extracting restaurant information according to the restaurant order information, performing attribute analysis, and analyzing the restaurant attribute of the target user in the current time period to obtain restaurant attribute analysis information; retrieving and acquiring characteristic information of various catering based on big data to form a catering data set; calculating the similarity between the catering attribute analysis information and the catering data set, judging the similarity with a preset threshold value, and classifying the catering tendency according to the judging result to obtain catering tendency analysis information; presetting a diet tendency analysis rule, and carrying out diet tendency analysis of a target user in a current time period by combining the diet tendency analysis information and the diet attribute analysis information; calculating the ratio of the number of various catering trends of the target user to the total number of catering trends, and comparing and analyzing with the eating trend analysis rule to obtain eating trend analysis information; obtaining user portrait information of a target user, including user preference portrait and user interest portrait, and carrying out diet preference change analysis according to the diet trend analysis information and the user portrait information to obtain diet preference change analysis information; personalized catering recommendation is carried out through the information of the change analysis of the dietary preference, the catering recommendation strategy and direction are dynamically adjusted in real time, and the method is close to the real-time preference and interest of the user;

It should be noted that, by analyzing the food order of the user in the last period, analyzing the food attribute in the order, judging whether the change occurs, obtaining the food attribute tendency of the target user in the last period, analyzing the diet tendency of the user according to the food attribute tendency mapping of the target user in the last period, analyzing the diet habit preference of the user, comparing and analyzing the diet tendency with the user portrait of the target user, judging whether the diet tendency change occurs, for example, changing from the previous food with high calorie preference to the food with low calorie preference, according to diet tendency change analysis information, the real-time tendency of the user can be known, and then the food close to the real-time situation is recommended, namely, the food with low calorie and high fiber is frequently selected in the current period of the user, the food selection conforming to the healthy diet target is recommended to the user, and more proper food is provided to meet the physical and health consideration of the user.

Fig. 3 is a block diagram 3 of a restaurant recommendation system based on user portraits according to an embodiment of the present invention, the system including: the memory 31 and the processor 32, wherein the memory 31 contains a user portrait based food and beverage recommendation method program, and the food and beverage recommendation method program based on the user portrait realizes the following steps when being executed by the processor 32:

It should be noted that the invention provides a restaurant recommendation method and a system based on user portraits, which analyze characteristics of a target user from the angles of user preference and user interest, and analyze interest points according to browsing behaviors of the user, so as to predict the interest points, including dining places, dining interests and dining time, and analyze proper restaurants by combining with current real-time weather, thereby giving more personalized experience to the user. Considering the influence of user preference and weather on the user, so as to provide catering suggestions which better meet the user's expectations and the current environment, and the personalized requirements of the user can be better met by integrating a plurality of information sources.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above described device embodiments are only illustrative, e.g. the division of the units is only one logical function division, and there may be other divisions in practice, such as: multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the various components shown or discussed may be coupled or directly coupled or communicatively coupled to each other via some interface, whether indirectly coupled or communicatively coupled to devices or units, whether electrically, mechanically, or otherwise.

The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units; can be located in one place or distributed to a plurality of network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated in one processing unit, or each unit may be separately used as one unit, or two or more units may be integrated in one unit; the integrated units may be implemented in hardware or in hardware plus software functional units.

Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, where the program, when executed, performs steps including the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, or the like, which can store program codes.

Alternatively, the above-described integrated units of the present invention may be stored in a computer-readable storage medium if implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in essence or a part contributing to the prior art in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, ROM, RAM, magnetic or optical disk, or other medium capable of storing program code.

The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A user portrayal-based catering recommendation method, comprising:

2. The restaurant recommendation method based on user portraits according to claim 1, wherein the attribute analysis is performed according to the user history order information to obtain attribute analysis result information, and the method specifically comprises the following steps:

3. The restaurant recommendation method based on user portraits according to claim 1, characterized in that said analyzing user preferences based on said user history order information and attribute analysis result information and constructing user preference portraits comprises:

4. The restaurant recommendation method based on user portraits according to claim 1, characterized in that said analyzing user interests based on said user history browsing information and constructing user interests portraits comprises:

5. A method of recommending dining based on user portraits as defined in claim 1, wherein said predicting the next point of interest for the target user based on the user interest portraits and the user preference portraits comprises:

6. The restaurant recommendation method based on user portraits as claimed in claim 1, wherein said obtaining real-time weather information, and combining user preference portraits, user interest portraits and point of interest prediction information, makes restaurant recommendation, specifically comprises:

7. A user portrayal-based dining recommendation system, the system comprising: the food and beverage recommendation method based on the user portrait comprises a memory and a processor, wherein the memory contains a food and beverage recommendation method program based on the user portrait, and when the food and beverage recommendation method program based on the user portrait is executed by the processor, the following steps are realized:

8. The user portrayal-based catering recommendation system according to claim 7, wherein the attribute analysis is performed according to the user history order information to obtain attribute analysis result information, and the method specifically comprises:

9. The restaurant recommendation system based on user portraits of claim 7, wherein said user preference analysis is performed based on said user history order information and attribute analysis result information and constructing user preference portraits comprises:

10. The food and beverage recommendation system based on user portrayal of claim 7, wherein said analyzing user interests based on said user history browsing information and constructing user interests portrayal comprises: