CN112232917A - Commodity recommendation method, device and equipment for e-commerce platform - Google Patents
Commodity recommendation method, device and equipment for e-commerce platform Download PDFInfo
- Publication number
- CN112232917A CN112232917A CN202011149837.4A CN202011149837A CN112232917A CN 112232917 A CN112232917 A CN 112232917A CN 202011149837 A CN202011149837 A CN 202011149837A CN 112232917 A CN112232917 A CN 112232917A
- Authority
- CN
- China
- Prior art keywords
- food material
- dishes
- user
- dish
- order information
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- 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
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
Abstract
The invention discloses a commodity recommendation method, a commodity recommendation device and commodity recommendation equipment for an e-commerce platform, wherein the method comprises the following steps: acquiring order information of each food material purchased by a user historically; determining that the environment of the position where the user purchases each food material order information is a historical environment and the dishes corresponding to each food material order information are alternative dishes; the method comprises the steps of obtaining the current environment of the current position of the user, determining the historical environment matched with the current environment as a target environment, determining alternative dishes corresponding to the same food material order information with the target environment as target dishes, and recommending corresponding food materials to the user based on the target dishes. According to the method and the device, the food materials are recommended according to dishes favored by the user in different environments, the recommended food materials can accord with the favorite of the user in different environments, and then targeted recommendation of the food materials to the user is achieved, accuracy in food material recommendation is improved, and user experience of the user is improved.
Description
Technical Field
The invention relates to the technical field of internet data processing, in particular to a commodity recommendation method, a commodity recommendation device and commodity recommendation equipment for an e-commerce platform.
Background
The community group purchase is a business mode based on development of deep ploughing communities, and is a mode with geographic attributes and local flow by integrating a small flow set of the communities; the customer acquisition cost is low, and the conversion rate and the retention rate are higher by adopting acquaintance sharing type marketing. But the community group buying mode is essentially retail, direct supply is directly realized at the source, the terminal contacts the consumer, the supply chain is long, the problems of any link are easy to form the bullwhip effect, and the subsequent link and the user experience are influenced. The community group-buying is mainly a group of users who are used to online shopping, do not have time to go off a retail store online, and a dish market realizes online buying behaviors, and when the community group-buying platform realizes commodity recommendation, food materials with activities or sales promotion and the like at present are generally recommended to each user directly, so that the food materials recommended by the realized food material recommendation can be food materials which are not needed by the user, or only part of the food materials are food materials needed by the user, and the user is required to select the food materials by himself or herself, so that the accuracy in the food material recommendation is low, and the user experience of the user is poor.
Disclosure of Invention
The invention aims to provide a commodity recommendation method, a commodity recommendation device and commodity recommendation equipment for an e-commerce platform, which can improve accuracy in food material recommendation and further improve user experience of a user.
In order to achieve the above purpose, the invention provides the following technical scheme:
a commodity recommendation method for an e-commerce platform comprises the following steps:
acquiring order information of each food material purchased by a user historically;
determining that the environment of the position where the user purchases each food material order information is a historical environment and the dishes corresponding to each food material order information are alternative dishes;
the method comprises the steps of obtaining the current environment of the current position of the user, determining the historical environment matched with the current environment as a target environment, determining alternative dishes corresponding to the same food material order information with the target environment as target dishes, and recommending corresponding food materials to the user based on the target dishes.
Preferably, the determining that the dishes corresponding to each food material order information are alternative dishes respectively includes:
determining any food material order information as current order information, and constructing a current food material vector based on food materials contained in the current order information;
respectively calculating the similarity between the current food material vector and each reference food material vector, and determining the dish corresponding to the reference food material vector with the similarity larger than the similarity threshold as an alternative dish; the reference food material vector is constructed based on food materials contained in any dish in each dish acquired in advance.
Preferably, the method further comprises the following steps:
if the reference food material vector with the similarity larger than the similarity threshold value does not exist, acquiring dish browsing information of the user in a preset time period of the receiving time of the current order information;
and respectively constructing to-be-selected food material vectors based on food materials contained in each dish in the dish browsing information, calculating the similarity between the current food material vector and each to-be-selected food material vector, and determining the dish corresponding to the to-be-selected food material vector with the highest similarity as an alternative dish.
Preferably, before calculating the similarity between different food material vectors, the method further includes:
preprocessing the food material vectors into vectors with consistent dimensions; the food material vector comprises a current food material vector, a reference food material vector and a to-be-selected food material vector.
Preferably, recommending a corresponding food material to the user based on the target dish includes:
determining that a target dish which is not made within a first preset time period and/or purchased with corresponding food materials within a second preset time period is a dish to be recommended, and recommending the food materials corresponding to the dish to be recommended to the user; the first preset time period and the second preset time period are both the time period before the current time and the time period closest to the current time, and the first preset time period is smaller than the second preset time period.
Preferably, after determining that the dishes corresponding to each piece of food material order information are alternative dishes, the method further includes:
classifying the alternative dishes to obtain a vegetable system to which each alternative dish belongs;
preferably, the method further comprises:
if the target dishes are all dishes which are made within a first preset time period and/or purchased with corresponding food materials within a second preset time period, determining that other dishes except the target dishes in the cuisine to which the target dishes belong are all current dishes, determining that the current dishes which are not made within the first preset time period and/or purchased with corresponding food materials within the second preset time period are dishes to be recommended, and recommending the food materials corresponding to the dishes to be recommended to the user.
Preferably, after determining that the dishes corresponding to each piece of food material order information are alternative dishes, the method further includes:
constructing a user representation of the user; the user portrait comprises a user identification of the user, a reference environment corresponding to the same food material order information and alternative dishes;
correspondingly, determining alternative dishes corresponding to the same food material order information with the target environment as the target dishes, including:
and determining the user identification containing the user and the user portrait of the target environment as a target portrait, and determining each alternative dish contained in the target portrait as a target dish.
A merchandise recommendation device for an e-commerce platform, comprising:
an acquisition module to: acquiring order information of each food material purchased by a user historically;
a determination module to: determining that the environment of the position where the user purchases each food material order information is a historical environment and the dishes corresponding to each food material order information are alternative dishes;
a recommendation module to: the method comprises the steps of obtaining the current environment of the current position of the user, determining the historical environment matched with the current environment as a target environment, determining alternative dishes corresponding to the same food material order information with the target environment as target dishes, and recommending corresponding food materials to the user based on the target dishes.
A merchandise recommendation device for an e-commerce platform, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the commodity recommendation method of the e-commerce platform as described in any one of the above when the computer program is executed.
The invention provides a commodity recommendation method, a commodity recommendation device and commodity recommendation equipment for an e-commerce platform, wherein the method comprises the following steps: acquiring order information of each food material purchased by a user historically; determining that the environment of the position where the user purchases each food material order information is a historical environment and the dishes corresponding to each food material order information are alternative dishes; the method comprises the steps of obtaining the current environment of the current position of the user, determining the historical environment matched with the current environment as a target environment, determining alternative dishes corresponding to the same food material order information with the target environment as target dishes, and recommending corresponding food materials to the user based on the target dishes. According to the technical scheme, the method comprises the steps of firstly obtaining an environment of a position where a user historically purchases food materials and dishes corresponding to the purchased food materials, then obtaining the environment of the current position of the user, determining the dishes corresponding to the historical environment matched with the environment of the current position of the user, possibly recommending the dishes which the user likes to eat under the current position of the user, and recommending the food materials corresponding to the dishes to the user; therefore, dish hobbies of the user in different environments are obtained based on the historical information of the food purchased by the user, and food materials required by corresponding dishes are recommended to the user based on the current environment where the user is located, so that the food materials are recommended according to the dish hobbies of the user in different environments, the recommended food materials can be matched with the hobbies of the user in different environments, targeted recommendation of the food materials to the user is achieved, accuracy in food material recommendation is improved, and user experience of the user is improved.
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 embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a commodity recommendation method for an e-commerce platform according to an embodiment of the present invention;
fig. 2 is a schematic diagram of order information generated by a user on a certain community group-buying platform on a certain day in the commodity recommendation method for the e-commerce platform according to the embodiment of the present invention;
fig. 3 is a schematic diagram of order information recommended to a user in a commodity recommendation method for an e-commerce platform according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a commodity recommendation device of an e-commerce platform according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a flowchart of a method for recommending a commodity on an e-commerce platform according to an embodiment of the present invention is shown, where the method includes:
s11: acquiring the order information of each food material purchased by a user historically.
The execution main body of the commodity recommendation method for the e-commerce platform provided by the embodiment of the invention can be a corresponding commodity recommendation device of the e-commerce platform; the user can be a user of the community group purchase platform, the user uses the community group purchase platform (namely an e-commerce platform) to purchase food materials of three meals a day, such as dinner, and the food materials are recommended as commodities of the community group purchase platform at the moment, so that the food material recommendation in the embodiment of the application is also commodity recommendation; the method includes acquiring food material order information of each food material purchased by a user historically, wherein the food material order information of the food material purchased by the user in the last year (or last n years, specifically, the information may be set according to actual needs), and the food material order information may include names, specifications, quantities, receiving time, information of positions of the food materials when the food materials are received, and the like.
S12: determining that the environment of the position where the user purchases each food material order information is a historical environment, and determining that the dishes corresponding to each food material order information are alternative dishes.
It should be noted that the information of the location of the food material order information at the time of receiving the goods, which is included in each food material order information, can be obtained, that is, the location of the user at the time of purchasing each food material order information; then, by combining the receiving time contained in each food material order information, the average daily air temperature and the time period (season, solar term and the like) corresponding to the receiving time and the area where the receiving position is located can be obtained; the daily average temperature and the time period can be used as the environment, so that the corresponding environment can be obtained based on the receiving position and the receiving time when the user purchases each food material order information, and the historical environment corresponding to the food material order information one by one is obtained. In addition, the food materials purchased by the user can be determined from any food material order information, and then dishes corresponding to the food material order information can be determined through the food materials, and the dishes can be called as alternative dishes.
S13: the method comprises the steps of obtaining the current environment of the current position of a user, determining the historical environment matched with the current environment as a target environment, determining alternative dishes corresponding to the same food material order information with the target environment as target dishes, and recommending corresponding food materials to the user based on the target dishes.
After the environment and the dishes of the position where the food material is purchased by the user every time in a short period of time are obtained, the user is considered to like eating the dishes corresponding to the food material order information under the environment of the position where the user purchases any food material order information, on the basis, the environment of the current position where the user is located (namely the environment of the current goods receiving position of the user) can be obtained, the historical environment matched with the environment of the current position where the user is located is inquired, the food material order information corresponding to the inquired historical environment is determined, and the dishes corresponding to the inquired food material order information are finally determined to be the dishes which the user prefers to eat under the current position environment. Wherein, the time-saving that two environment phase-matchings include two environments is the same, and the average temperature per day is more close (the difference of average temperature per day is in the certain limit that sets up according to actual need).
According to the technical scheme, the method comprises the steps of firstly obtaining an environment of a position where a user historically purchases food materials and dishes corresponding to the purchased food materials, then obtaining the environment of the current position of the user, determining the dishes corresponding to the historical environment matched with the environment of the current position of the user, possibly recommending the dishes which the user likes to eat under the current position of the user, and recommending the food materials corresponding to the dishes to the user; therefore, dish hobbies of the user in different environments are obtained based on the historical information of the food purchased by the user, and food materials required by corresponding dishes are recommended to the user based on the current environment where the user is located, so that the food materials are recommended according to the dish hobbies of the user in different environments, the recommended food materials can be matched with the hobbies of the user in different environments, targeted recommendation of the food materials to the user is achieved, accuracy in food material recommendation is improved, and user experience of the user is improved.
The commodity recommendation method for the e-commerce platform provided by the embodiment of the invention determines that dishes corresponding to each food material order information are alternative dishes respectively, and may include:
determining any food material order information as current order information, and constructing a current food material vector based on food materials contained in the current order information;
respectively calculating the similarity between the current food material vector and each reference food material vector, and determining the dish corresponding to the reference food material vector with the similarity larger than the similarity threshold as an alternative dish; the reference food material vector is constructed based on food materials contained in any dish in each dish acquired in advance.
When determining the corresponding dishes based on the food material order information, a food material vector corresponding to each food material order information may be constructed first, specifically, for any food material order information, various food materials contained in the food material order information can be arranged according to the sequence of main materials, auxiliary materials and ingredients, wherein the main material is the main food material for making dishes, for example, the main material of spicy crayfish is crayfish, the main material of stewed chicken is chicken, the auxiliary material is the auxiliary material for making dishes, for example, the auxiliary material in the spicy crayfish can be onion, the auxiliary material for stewing chicken can be potato, the auxiliary material is the seasoning for preparing dishes, such as light soy sauce, vinegar, etc., for example, the order information generated by a certain user on a certain community group buying platform on a certain day is as shown in fig. 2, and the food material vector constructed according to the order information is [ chicken breast, hen egg, red pepper, green Chinese onion, water chestnut ].
After food material vectors corresponding to each piece of food material order information are obtained, a menu food material vector set stored in a server can be called based on main materials and/or auxiliary materials in the food material vectors, wherein the menu food material vector set comprises a plurality of food material vectors (namely reference food material vectors), each food material vector is constructed by corresponding dishes in a menu, and therefore the food material vectors in the menu food material vector set correspond to the corresponding dishes one by one; calculating the similarity between each reference food material vector and the current food material vector, for example, the reference food material vector set of the spicy diced chicken is as follows:
{
[ chicken breast, egg, water chestnut, pickled pepper, green Chinese onion, ginger and garlic ],
[ chicken meat, egg white, red pepper, green onion, ginger, garlic, pea ],
[ Chicken breast, Green and Red Pepper, Lentinus edodes, dried Capsici fructus, pickled Pepper ]
}
It should be noted that the similarity between each reference food material vector and the current food material vector can be calculated by using a similarity calculation method, where the similarity calculation method includes, but is not limited to, an euclidean distance, a manhattan distance, a chebyshev distance, a minkowski distance, a normalized euclidean distance, a mahalanobis distance, an included angle cosine, a hamming distance, a jaka distance, and the like, and the similarity between the current food material vector and the reference food material vector can be calculated by using the included angle remainder similarity according to the following formula:
wherein A represents a current food material vector, B represents a reference food material vector, AiRepresenting the food material elements (i.e., the food materials) in the current food material vector, BiRepresenting the food material elements in the reference food material vector.
After obtaining the similarity between the current food material vector and each reference food material vector, if the similarity is greater than a similarity threshold (e.g., 0.6) set according to actual needs, it may be determined that the reference food material vector corresponding to the similarity greater than the similarity threshold is the reference food material vector matched with the current food material vector, and the dish of the reference food material vector is the dish corresponding to the current food material vector, that is, the dish of the food material order information corresponding to the current food material vector.
Therefore, the food material vectors of the known corresponding dishes with higher similarity to the food material vectors constructed by the food materials contained in the food material order information are determined, and the determined dishes are dishes corresponding to the food material order information, so that the dishes corresponding to the food material order information are effectively determined.
The commodity recommendation method for the e-commerce platform provided by the embodiment of the invention can further comprise the following steps:
if the reference food material vector with the similarity larger than the similarity threshold value does not exist, acquiring dish browsing information of a user in a preset time period of the receiving time of the current order information;
and respectively constructing to-be-selected food material vectors based on food materials contained in each dish in the dish browsing information, calculating the similarity between the current food material vector and each to-be-selected food material vector, and determining the dish corresponding to the to-be-selected food material vector with the highest similarity as an alternative dish.
If the similarity between the reference food material vector and the current food material vector is not greater than the similarity threshold, determining the dishes corresponding to the current food material vector based on the historical records of the user for making the dishes; specifically, the dish browsing information of the user in a preset time period (for example, in each week before and after, specifically, the dish browsing information may be set according to actual needs) of the harvest time of the current order information may be obtained, the dish browsing information may include one or more data browsing records of a video browsing record, a character browsing record and a picture browsing record of the user in a corresponding time period, the video browsing record is a record of a video of the user browsing the dish, the character browsing record is a record of a character of the user browsing the dish, the picture browsing record is a record of a picture of the dish browsed by the user, and the content browsed by the user when the user browses the dish may be a menu corresponding to the dish, specifically, the content may include food materials (including food material name, weight and the like) required by the dish, a dish making method, a dish nutritional value and the like, so that each dish browsed by the user may be determined based on the dish browsing information, and constructing a food material vector constructed by each food material required by the manufacture of each dish in the dishes, namely obtaining the food material vectors to be selected which have one-to-one correspondence with the dishes, calculating the similarity between each food material vector to be selected and the current food material vector, and finally determining that the food material vector to be selected corresponding to the highest similarity is the food material vector to be selected which is matched with the current food material vector, and the dish of the food material vector to be selected is the dish corresponding to the current food material vector, namely the dish of the food material order information corresponding to the current food material vector. For example, a user browses dish related information (such as a menu) of fish-flavor shredded pork, chicken in mouth, spicy crawfish, spicy diced chicken, fried shrimps and the like in a week, and judges one or more food material elements containing main food materials in a current food material vector of the user [ chicken breast, hen eggs, red pepper, green Chinese onion and water chestnut ], the spicy diced chicken is processed with a high probability after the receiving time, and therefore the dish corresponding to the current food material vector is the spicy diced chicken.
Therefore, the dishes browsed by the user in the period of time are determined by the dish browsing information of the user in the certain time range of the food material order information receiving time, and then the dishes which are closest to the food material and are contained in the food material order information are determined from the dishes, so that after the food material order information contains various food materials, the dishes which are most probably made are also the dishes corresponding to the food material order information, and therefore effective determination of the dishes corresponding to the food material order information is achieved.
Before calculating the similarity between different food material vectors, the commodity recommendation method for the e-commerce platform provided by the embodiment of the invention may further include:
preprocessing the food material vectors into vectors with consistent dimensions; the food material vector comprises a current food material vector, a reference food material vector and a to-be-selected food material vector.
In order to facilitate calculation of the similarity between the food material vectors, the food material vectors can be subjected to uniform preprocessing based on the main material and the auxiliary material before the similarity between the food material vectors is calculated, so that the dimensions of the two food material vectors participating in the similarity calculation are consistent; the method comprises the steps that when food material vectors are preprocessed in a statistical mode based on main materials and auxiliary materials, ingredients, auxiliary materials and the main materials are sequentially arranged from high to low in priority, then one food material vector with more food material elements is contained in the two food material vectors participating in similarity calculation, corresponding food material elements are deleted according to the priority from high to low until the number of the food material elements contained in the two food material vectors is the same, so that the food material elements with the smallest influence on the food material vector corresponding to dishes are determined preferentially to be deleted, and the accuracy of the similarity calculation of the two food material vectors is guaranteed.
After determining the dishes corresponding to each food material order information, the dishes can be classified to obtain the cuisine to which each dish belongs, for example, the cuisine comprises Mapo bean curd, spicy crayfish, spicy hot pot, cooked pork, boiled chicken, lung slices of a couple and the like, so that the corresponding cuisine can be determined to be Sichuan dish; of course, one user may correspond to more than one cuisine.
In addition, after the target dish and the dish system to which the target dish belongs are obtained, dishes conforming to the favorite style of the target dish and food materials corresponding to the dishes can be recommended to the user based on the target dish, the dish system to which the target dish belongs and recent dish browsing information of the user, and specifically, after the target dish is determined, various food materials contained in the target dish can be recommended to the user; in practice, if the same dishes are frequently eaten, the user generally does not want to eat repeated dishes in a short period, and corresponding judgment can be carried out through the frequency of the user purchasing corresponding food materials of the dishes and/or the frequency of the user making the dishes on the basis of the method and the device for eating the dishes; meanwhile, the target dishes are directly determined based on the food material purchase information of the user, so that the target dishes are generally favored by the user compared with other dishes in the same dish system, and therefore the target dishes are preferentially selected when the food material recommendation is realized in the application, and then the other dishes in the same dish system are considered.
When the target dish is preferentially selected to realize food material recommendation, the commodity recommendation method of the e-commerce platform provided by the embodiment of the invention can recommend a corresponding food material to a user based on the target dish, and can include:
determining that a target dish which is not made within a first preset time period and/or purchased with corresponding food materials within a second preset time period is a dish to be recommended, and recommending the food materials corresponding to the dish to be recommended to a user; the first preset time period and the second preset time period are both the time period before the current moment and the time period nearest to the current moment, and the first preset time period is smaller than the second preset time period.
The first preset time period and the second preset time period are parameters representing frequency of making corresponding dishes by a user and parameters representing food materials corresponding to the dishes purchased by the user, and both the first preset time period and the second preset time period can be set according to actual needs, for example, the first preset time period is about 3 days, and the second preset time period is about 1 week. When the food materials required by the dishes are recommended to the user, the food materials corresponding to the recommended target dishes are considered, specifically, if any dish has been made in the near future and/or any food material corresponding to the dish has been purchased in the near future, the food materials corresponding to the dish do not need to be recommended to the user in order to avoid the user from eating the same dish repeatedly in a short period. Therefore, the situation that the user repeatedly makes the same dishes in a short time is avoided, and the recommendation of the corresponding food materials is preferentially realized based on the target dishes, so that the current requirements of the user are met by the food material recommendation.
When the target dish does not meet the requirement and other dishes in the same family are selected to realize food material recommendation, the commodity recommendation method of the e-commerce platform provided by the embodiment of the invention can further include the following steps after determining that the dish corresponding to each food material order information is an alternative dish respectively:
classifying the alternative dishes to obtain a vegetable system to which each alternative dish belongs;
correspondingly, the method may further include:
if the target dishes are all dishes which are made within a first preset time period and/or purchased with corresponding food materials within a second preset time period, determining that other dishes except the target dishes in the family to which the target dishes belong are all current dishes, determining that the current dishes which are not made within the first preset time period and/or purchased with corresponding food materials within the second preset time period are dishes to be recommended, and recommending the food materials corresponding to the dishes to be recommended to the user.
If all target dishes are dishes which are made or purchased with corresponding food materials recently, recommendation of food materials corresponding to other dishes of the same family can be considered, specifically, the other dishes of the same family are used as current dishes, and then recommendation of the food materials of the other dishes of the same family is achieved according to the principle when the target dishes are recommended. In addition, if other dishes of the same cuisine are all made or corresponding food materials are purchased recently, the dish with the time for purchasing the corresponding food material farthest from the current time or the time for making the corresponding food material farthest from the current time can be selected from the target dish and the other dishes of the same cuisine as the dish to be recommended, and then recommendation of the corresponding food material is achieved.
The steps of recommending food materials are illustrated, for example, if the dish is a spicy crayfish, a single dish recommending frequency can be set, for example, if the user has purchased a crayfish in the last week, the user can be judged whether the user has done the spicy crayfish or not by combining dish browsing information when the user browses the dish, and after judgment, the user may have done the spicy crayfish on the current week, and if the single dish recommending frequency is 3 days, the related food materials for making the spicy crayfish should not be recommended within the 3 days; for example, if the user does not purchase chicken within a week, it may be determined that the user has not done any dish related to chicken within a week, and some food materials required for the chicken practice in daily life, such as recommended chicken-in-mouth food materials, may be recommended to the user. Of course, other arrangements according to actual needs are within the protection scope of the present invention.
According to the method and the device, when whether the user has made a certain dish or not is judged, dish browsing information of data corresponding to the dish in time after the user has purchased food materials corresponding to the dish for the last time can be calculated, the favorite score of the dish is calculated based on the dish browsing information, if the favorite score reaches a threshold value set according to actual needs, the user is considered to have made the dish, and otherwise, the user is considered not to have made the dish. The calculating of the preference score may specifically include:
for any dish in each item contained in the dish browsing information, acquiring time length information and a time period in which the user browses corresponding information of the dish, and calculating the favorite score of the dish based on the time length information and the time period;
calculating the favorite score of the current dish based on the time length information and the time period may include:
acquiring a time length score corresponding to the time length information and a time period score corresponding to the time period; based on weighted values preset for the duration information and the time period, carrying out weighted summation on the duration score and the time period score to obtain a favorite score of the dish; the duration information comprises duration for browsing the information corresponding to the dishes or the proportion of the viewed part in all the information when the information corresponding to the dishes is browsed, and the longer the duration or the larger the proportion is, the higher the duration score is, the closer the time period is to the dining time, and the higher the time period score is.
It should be noted that the favorite style of dishes in a family is generally consistent, and the overall style is not changed greatly; because the cooking time of dinner is generally concentrated in the time period of 17:00 to 19:00, a main consumer group of a plurality of users, particularly a community group purchase platform, is a young person, and the main consumer group likes to cook according to a menu, and usually cooks by browsing open menus such as videos, webpages (including characters, pictures and the like) and the like; the operation behaviors of the user for browsing the information generally search by taking food materials or names of dishes to be made as key words, then select a menu in the form of a suitable dish making video, characters, pictures or the like based on the search result, and make the dishes according to the dishes, so that the favorite style of the family where the user is located is judged to be easier generally in a time period from 17:00 to 19:00, and the dishes made by the user can be judged by extracting the browsing information of the user in the corresponding time period, and similarly, the favorite style of the family where the user is located is judged to be easier in other dining times; based on the method and the device, 24 hours in a day can be divided into a plurality of time periods in advance, and then a corresponding time period score is set for each time period based on the easiness of obtaining the favorite style of dishes of the family where the user is located, namely, the more easily the favorite style of the dishes of the family where the user is located is obtained (or the closer the time for the user to browse the corresponding dishes is to the dining time), the higher the corresponding time period score is; for the convenience of calculating the present application, the time period scores may be normalized so that the sum of the time period scores is 1, and in a specific example, the time periods and the time period scores of the time periods may be as shown in table 1. It should be noted that, for the duration information, when the browsed data is a video, the duration information may include a ratio of a duration of watching the video to a duration of the entire video, and when the browsed data is a text or a picture, the duration information may include a duration of watching the text or the picture, and the larger the ratio in the duration information and the longer the duration are, the more interested the user is in the current dish is indicated, so that corresponding duration scores may be set for different duration information according to this manner, that is, the larger the ratio in the duration information is or the longer the duration is, the higher the duration score is, and for the convenience of calculation, the time period scores may be normalized so that the sum of the time period scores is 1.
In addition, the application can also set corresponding weighted values for the time period score and the time length score respectively, and the sum of the weighted values of the time period score and the time length score is 1. Based on the setting, after the time period of the dish and the duration information of the dish are obtained when the user browses the dish, the time period fraction of the time period of the dish is multiplied by the weighted value of the time period fraction to obtain a corresponding product, the duration fraction of the duration information of the dish is multiplied by the weighted value of the duration fraction to obtain a corresponding product, finally the two products are added to obtain the favorite fraction of the dish, and the dish with the favorite fraction larger than the fraction threshold value set according to actual needs is determined to be the dish made by the user. Therefore, the dishes made by the user can be effectively determined based on the duration information and the time period when the user browses the dishes.
TABLE 1
| Time period sequence number | Time of day | Time period fraction |
| 1 | 00:00-5:00 | 0.05 |
| 2 | 5:01-8:00 | 0.1 |
| 3 | 8:01-11:00 | 0.05 |
| 4 | 11:01-13:00 | 0.1 |
| 5 | 13:01-17:00 | 0.1 |
| 6 | 17:01-20:00 | 0.5 |
| 7 | 20:01-24:00 | 0.1 |
The method for recommending the e-commerce platform provided by the embodiment of the invention can further comprise the following steps of after determining that the dishes corresponding to each food material order information are alternative dishes respectively:
constructing a user image of a user; the user picture comprises a user identification of the user, a reference environment corresponding to the same food material order information and alternative dishes;
correspondingly, determining alternative dishes corresponding to the same food material order information with the target environment as the target dishes, including:
and determining a user portrait containing a user identification of the user and a target environment as a target portrait, and determining each alternative dish contained in the target portrait as a target dish.
The User identification can be an identification which uniquely represents a corresponding User, such as a User ID, a User number and the like, after the User identification and an environment and dishes of food materials purchased by the User in history are obtained, a User portrait can be obtained, namely (the User ID, a time period, a daily average air temperature and dishes), such as (the User ID, autumn, the daily average air temperature of 30 degrees and spicy diced chicken), and further when the target dish is determined, each dish contained in the User portrait containing the User identification and the target environment can be directly determined as the target dish, for example, (User _1124221, summer, 30-35 degrees centigrade, white gourd chop | tomato egg chicken soup | mashed garlic water spinach | … …) in the User portrait of plum four, so that when the temperature is 30-35 degrees centigrade in summer, the corresponding food material can be recommended to the User; in this way, management and query of the corresponding information are facilitated.
In conclusion, the personalized purchase recommendation can be performed based on the user preference, so that the food material recommendation function is provided for the community group purchase platform, namely the platform can realize food material recommendation through the technical scheme of the invention, the trouble that the user purchases and eats what dishes every day is avoided, and the SKU of the community group purchase platform can be improved.
According to the commodity recommendation method for the E-commerce platform provided by the embodiment of the invention, when recommending food materials to a user, various food materials needing to be recommended to the user can be recommended to the user in a list form, the list is purchase information of corresponding food materials and can comprise food material names, duration quantity or weight, food material prices and the like, if the recommended dish is a chicken, the corresponding purchase information can be as shown in fig. 3, so that the user can directly order the food materials based on the list or modify the information in the list and the like, the purchase of the recommended food materials is facilitated for the user, and the user experience of the user is further improved.
According to the commodity recommendation method for the E-commerce platform, provided by the embodiment of the invention, the food materials are recommended to the user, and the menu of dishes (including the dish making method) can be recommended to the user, so that the user can know the dish making method while knowing each recommended food material, the user can conveniently realize the operations of purchasing the food materials, making the dishes and the like based on the operation, and the user experience of the user is further improved.
An embodiment of the present invention further provides a commodity recommendation device for an e-commerce platform, as shown in fig. 4, the commodity recommendation device may include:
an obtaining module 11, configured to: acquiring order information of each food material purchased by a user historically;
a determining module 12 for: determining that the environment of the position where the user purchases each food material order information is a historical environment and dishes corresponding to each food material order information are alternative dishes;
a recommendation module 13 configured to: the method comprises the steps of obtaining the current environment of the current position of a user, determining the historical environment matched with the current environment as a target environment, determining alternative dishes corresponding to the same food material order information with the target environment as target dishes, and recommending corresponding food materials to the user based on the target dishes.
The commodity recommendation device for the e-commerce platform provided by the embodiment of the invention comprises a determining module and a recommending module, wherein the determining module comprises:
a first determination unit configured to: determining any food material order information as current order information, and constructing a current food material vector based on food materials contained in the current order information; respectively calculating the similarity between the current food material vector and each reference food material vector, and determining the dish corresponding to the reference food material vector with the similarity larger than the similarity threshold as an alternative dish; the reference food material vector is constructed based on food materials contained in any dish in each dish acquired in advance.
The commodity recommendation device for the e-commerce platform provided by the embodiment of the invention further comprises a determining module, wherein the determining module comprises:
a second determination unit configured to: if the reference food material vector with the similarity larger than the similarity threshold value does not exist, acquiring dish browsing information of a user in a preset time period of the receiving time of the current order information; and respectively constructing to-be-selected food material vectors based on food materials contained in each dish in the dish browsing information, calculating the similarity between the current food material vector and each to-be-selected food material vector, and determining the dish corresponding to the to-be-selected food material vector with the highest similarity as an alternative dish.
The commodity recommendation device for the e-commerce platform provided by the embodiment of the invention can further comprise:
a pre-processing module to: preprocessing the food material vectors into vectors with consistent dimensions before calculating the similarity among different food material vectors; the food material vector comprises a current food material vector, a reference food material vector and a to-be-selected food material vector.
The embodiment of the invention provides a commodity recommendation device of an e-commerce platform, and a recommendation module comprises:
a first recommending unit configured to: determining that a target dish which is not made within a first preset time period and/or purchased with corresponding food materials within a second preset time period is a dish to be recommended, and recommending the food materials corresponding to the dish to be recommended to a user; the first preset time period and the second preset time period are both the time period before the current moment and the time period nearest to the current moment, and the first preset time period is smaller than the second preset time period.
The commodity recommendation device for the e-commerce platform provided by the embodiment of the invention can further comprise:
a classification module to: after determining that the dishes corresponding to each food material order information are alternative dishes respectively, classifying the alternative dishes to obtain a dish system to which each alternative dish belongs;
correspondingly, the recommending module can further comprise:
a second recommending unit for: if the target dishes are all dishes which are made within a first preset time period and/or purchased with corresponding food materials within a second preset time period, determining that other dishes except the target dishes in the family to which the target dishes belong are all current dishes, determining that the current dishes which are not made within the first preset time period and/or purchased with corresponding food materials within the second preset time period are dishes to be recommended, and recommending the food materials corresponding to the dishes to be recommended to the user.
The commodity recommendation device for the e-commerce platform provided by the embodiment of the invention can further comprise:
a build module to: after determining that the dishes corresponding to each food material order information are alternative dishes respectively, constructing a user image of the user; the user picture comprises a user identification of the user, a reference environment corresponding to the same food material order information and alternative dishes;
correspondingly, the recommending module can include:
a third determination unit configured to: and determining a user portrait containing a user identification of the user and a target environment as a target portrait, and determining each alternative dish contained in the target portrait as a target dish.
The embodiment of the invention also provides commodity recommendation equipment of the E-commerce platform, which comprises the following components:
a memory for storing a computer program;
and the processor is used for realizing the steps of the commodity recommendation method of the E-commerce platform when executing the computer program.
It should be noted that for the description of the relevant parts in the commodity recommendation device, the equipment and the storage medium of the e-commerce platform provided in the embodiment of the present invention, reference is made to the detailed description of the corresponding parts in the commodity recommendation method of the e-commerce platform provided in the embodiment of the present invention, and details are not repeated here. In addition, parts of the technical solutions provided in the embodiments of the present invention that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail, so as to avoid redundant description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A commodity recommendation method for an E-commerce platform is characterized by comprising the following steps:
acquiring order information of each food material purchased by a user historically;
determining that the environment of the position where the user purchases each food material order information is a historical environment and the dishes corresponding to each food material order information are alternative dishes;
the method comprises the steps of obtaining the current environment of the current position of the user, determining the historical environment matched with the current environment as a target environment, determining alternative dishes corresponding to the same food material order information with the target environment as target dishes, and recommending corresponding food materials to the user based on the target dishes.
2. The method of claim 1, wherein determining that the dishes corresponding to each food material order information are alternative dishes respectively comprises:
determining any food material order information as current order information, and constructing a current food material vector based on food materials contained in the current order information;
respectively calculating the similarity between the current food material vector and each reference food material vector, and determining the dish corresponding to the reference food material vector with the similarity larger than the similarity threshold as an alternative dish; the reference food material vector is constructed based on food materials contained in any dish in each dish acquired in advance.
3. The method of claim 2, further comprising:
if the reference food material vector with the similarity larger than the similarity threshold value does not exist, acquiring dish browsing information of the user in a preset time period of the receiving time of the current order information;
and respectively constructing to-be-selected food material vectors based on food materials contained in each dish in the dish browsing information, calculating the similarity between the current food material vector and each to-be-selected food material vector, and determining the dish corresponding to the to-be-selected food material vector with the highest similarity as an alternative dish.
4. The method of claim 3, wherein before calculating the similarity between different food material vectors, further comprising:
preprocessing the food material vectors into vectors with consistent dimensions; the food material vector comprises a current food material vector, a reference food material vector and a to-be-selected food material vector.
5. The method of claim 1, wherein recommending respective food materials to the user based on the target dish comprises:
determining that a target dish which is not made within a first preset time period and/or purchased with corresponding food materials within a second preset time period is a dish to be recommended, and recommending the food materials corresponding to the dish to be recommended to the user; the first preset time period and the second preset time period are both the time period before the current time and the time period closest to the current time, and the first preset time period is smaller than the second preset time period.
6. The method of claim 5, wherein after determining that the dish corresponding to each piece of food material order information is a candidate dish, the method further comprises:
and classifying the alternative dishes to obtain the vegetable series to which each alternative dish belongs.
7. The method of claim 6, further comprising:
if the target dishes are all dishes which are made within a first preset time period and/or purchased with corresponding food materials within a second preset time period, determining that other dishes except the target dishes in the cuisine to which the target dishes belong are all current dishes, determining that the current dishes which are not made within the first preset time period and/or purchased with corresponding food materials within the second preset time period are dishes to be recommended, and recommending the food materials corresponding to the dishes to be recommended to the user.
8. The method of claim 1, wherein after determining that the dish corresponding to each piece of food material order information is a candidate dish, the method further comprises:
constructing a user representation of the user; the user portrait comprises a user identification of the user, a reference environment corresponding to the same food material order information and alternative dishes;
correspondingly, determining alternative dishes corresponding to the same food material order information with the target environment as the target dishes, including:
and determining the user identification containing the user and the user portrait of the target environment as a target portrait, and determining each alternative dish contained in the target portrait as a target dish.
9. A merchandise recommendation device for an e-commerce platform, comprising:
an acquisition module to: acquiring order information of each food material purchased by a user historically;
a determination module to: determining that the environment of the position where the user purchases each food material order information is a historical environment and the dishes corresponding to each food material order information are alternative dishes;
a recommendation module to: the method comprises the steps of obtaining the current environment of the current position of the user, determining the historical environment matched with the current environment as a target environment, determining alternative dishes corresponding to the same food material order information with the target environment as target dishes, and recommending corresponding food materials to the user based on the target dishes.
10. An article recommendation device for an e-commerce platform, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the method for recommending merchandise for an e-commerce platform according to any one of claims 1 to 8 when executing said computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011149837.4A CN112232917B (en) | 2020-10-23 | 2020-10-23 | Commodity recommendation method, device and equipment for e-commerce platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011149837.4A CN112232917B (en) | 2020-10-23 | 2020-10-23 | Commodity recommendation method, device and equipment for e-commerce platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112232917A true CN112232917A (en) | 2021-01-15 |
| CN112232917B CN112232917B (en) | 2022-05-13 |
Family
ID=74110328
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011149837.4A Active CN112232917B (en) | 2020-10-23 | 2020-10-23 | Commodity recommendation method, device and equipment for e-commerce platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112232917B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335365A (en) * | 2014-05-30 | 2016-02-17 | 小米科技有限责任公司 | Method and apparatus for placing order, server and terminal |
| CN107742245A (en) * | 2017-10-31 | 2018-02-27 | 北京小度信息科技有限公司 | A kind of merchant information recommends method, apparatus and equipment |
| CN109119142A (en) * | 2018-08-30 | 2019-01-01 | 周梦杰 | A kind of personalization preferred diet appraisal procedure |
| CN109146630A (en) * | 2018-08-20 | 2019-01-04 | 上海掌门科技有限公司 | Information-pushing method |
| CN110335118A (en) * | 2019-07-04 | 2019-10-15 | 合肥美的电冰箱有限公司 | Menu recommended method, menu recommendation apparatus and machine readable storage medium |
| CN110766494A (en) * | 2018-07-27 | 2020-02-07 | 博西华电器（江苏）有限公司 | List generation method, device and system |
| CN111402013A (en) * | 2020-06-04 | 2020-07-10 | 成都晓多科技有限公司 | Commodity collocation recommendation method, system, device and storage medium |
| WO2020156403A1 (en) * | 2019-02-01 | 2020-08-06 | 上海掌门科技有限公司 | Method and device used for generating order |
-
2020
- 2020-10-23 CN CN202011149837.4A patent/CN112232917B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335365A (en) * | 2014-05-30 | 2016-02-17 | 小米科技有限责任公司 | Method and apparatus for placing order, server and terminal |
| CN107742245A (en) * | 2017-10-31 | 2018-02-27 | 北京小度信息科技有限公司 | A kind of merchant information recommends method, apparatus and equipment |
| CN110766494A (en) * | 2018-07-27 | 2020-02-07 | 博西华电器（江苏）有限公司 | List generation method, device and system |
| CN109146630A (en) * | 2018-08-20 | 2019-01-04 | 上海掌门科技有限公司 | Information-pushing method |
| CN109119142A (en) * | 2018-08-30 | 2019-01-01 | 周梦杰 | A kind of personalization preferred diet appraisal procedure |
| WO2020156403A1 (en) * | 2019-02-01 | 2020-08-06 | 上海掌门科技有限公司 | Method and device used for generating order |
| CN110335118A (en) * | 2019-07-04 | 2019-10-15 | 合肥美的电冰箱有限公司 | Menu recommended method, menu recommendation apparatus and machine readable storage medium |
| CN111402013A (en) * | 2020-06-04 | 2020-07-10 | 成都晓多科技有限公司 | Commodity collocation recommendation method, system, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112232917B (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11669557B2 (en) | Iterative image search algorithm informed by continuous human-machine input feedback | |
| US9639805B1 (en) | Inferring temporal attributes of a recipe | |
| Miller et al. | Nutrition and cost comparisons of select canned, frozen, and fresh fruits and vegetables | |
| CA2838084A1 (en) | System and method for providing flavor advisement and enhancement | |
| CN103858142A (en) | Store information provision system | |
| JP6903523B2 (en) | Information processing equipment, information processing methods, and programs | |
| JP5751883B2 (en) | Menu proposal system, method and program thereof | |
| Luckett et al. | Consumer attitudes toward texture and other food attributes | |
| CN109214956B (en) | Meal pushing method and device | |
| JP6410069B1 (en) | Recipe information providing apparatus, recipe information providing method, and recipe information providing program | |
| JP2019133624A (en) | Recipe information provision apparatus, recipe information provision method, and recipe information provision program | |
| CN112232917B (en) | Commodity recommendation method, device and equipment for e-commerce platform | |
| Hernández Medina et al. | Culture and healthy lifestyle: factors influencing the decision to buy quinoa in the City of Latacunga in Cotopaxi Province, Ecuador | |
| JP2019191626A (en) | Shopping supporting system, shopping supporting server, program, and user terminal | |
| CN115221420A (en) | Diet recommendation method and system based on user portrait | |
| CN112232916B (en) | Commodity recommendation method, device and equipment | |
| Ferdouse | Tilapia in Asian Markets... can we sell more? | |
| Qiao et al. | Privacy‐preserving dish‐recommendation for food nutrition through edging computing | |
| Kato et al. | Sizzle word analysis from multiple internet media | |
| JP2023154330A (en) | Shopping support system, shopping support method, and program | |
| Ajayi et al. | Application of mobile phones in the marketing of banana in Ondo State, Nigeria. | |
| Love et al. | Nutrition and origin of United States chain restaurant seafood | |
| Veit | American Food, Cooking, and Nutrition, 1900–1945 | |
| Azman et al. | CONSUMER BEHAVIOUR: READY-TO-EAT FOOD IN PENINSULAR MALAYSIA BASED ON FOOD-RELATED LIFESTYLE | |
| CN117174253A (en) | Personalized ketogenic diet recommendation method and system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |