CN110727858A - Recommendation method, computer storage medium and electronic device - Google Patents
Recommendation method, computer storage medium and electronic device Download PDFInfo
- Publication number
- CN110727858A CN110727858A CN201910846332.4A CN201910846332A CN110727858A CN 110727858 A CN110727858 A CN 110727858A CN 201910846332 A CN201910846332 A CN 201910846332A CN 110727858 A CN110727858 A CN 110727858A
- Authority
- CN
- China
- Prior art keywords
- user
- class
- weight
- weights
- representation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Entrepreneurship & Innovation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A recommendation method, a computer storage medium and an electronic device include: acquiring a first user portrait of each stored user and a second user portrait of a user to be recommended; the user representation includes a classification; determining a first weight of each type in each first user representation and a second weight of each type in the second user representation; determining a first user representation that is most similar to the second user representation among all of the first user representations based on the first weight and the second weight; and recommending the user corresponding to the most similar first user portrait to the user to be recommended. According to the scheme, various first weights in each first user portrait and various second weights in each second user portrait are determined, and then similar users are determined according to the weights.
Description
Technical Field
The present application relates to mobile communication technologies, and in particular, to a recommendation method, a computer storage medium, and an electronic device.
Background
In recent years, the logistics industry is increasingly competitive, the largest problem of the logistics industry is matching of vehicles and goods, and the improvement of the matching success rate depends on the relationship between the two parties of the vehicles and the goods to a great extent. The logistics people need to continuously carry out stranger social contact to identify the logistics people mutually matched with the self requirements, so that potential business opportunities are mined.
Traditional offline social contact has few social opportunities and low efficiency. Many online logistics platforms also come up with some functionality for recommending users. But these functions generally query other users who have published compliant requirements for recommendations according to the requirements that the users published in their platforms. This recommendation, although straightforward, has some drawbacks:
1. the logic formed by recommendation is single, and the social contact among people is generated by giving a plurality of complex factors, so that the social contact requirement of the user beyond the working requirement of the user cannot be reflected
2. The logistics requirements often have certain timeliness, and the published requirements can not meet the current social attributes of the users along with the changes of work posts, processes and environments of the users.
Therefore, the accuracy of such recommendations is low.
Disclosure of Invention
The embodiment of the application provides a recommendation method, a computer storage medium and an electronic device, so as to solve the technical problems.
According to a first aspect of embodiments of the present application, there is provided a recommendation method, including:
acquiring a first user portrait of each stored user and a second user portrait of a user to be recommended; the user representation includes a classification;
determining a first weight of each type in each first user representation and a second weight of each type in the second user representation;
determining a first user representation that is most similar to the second user representation among all first user representations based on the first and second weights;
and recommending the user corresponding to the most similar first user portrait to the user to be recommended.
Optionally, the classification of the user representation comprises: basic attribute class, company attribute class, interest and hobby class, historical service information class and user behavior information class;
the basic attribute class, the company attribute class, the interest and hobby class, the historical service information class and the user behavior information class all have initial weights and the latest updating dates of the initial weights;
the initial weight of the basic attribute class is 3, the initial weight of the company attribute class is 2, the initial weight of the interest and hobby class is 2, the initial weight of the historical service information class is 2, and the initial weight of the user behavior information class is 1.
Optionally, the basic attribute class includes at least one or more of the following attributes: age, working age, location of mobile phone number;
the company attributes include at least one or more of the following attributes: company business range, company scale, province of company business, company partner;
the interest taste class includes at least one or more of the following attributes: personal interest points, industry categories of interest, business categories of interest, professional categories of interest;
the historical service information class includes at least one or more of the following attributes: service type, service occurrence time and service occurrence place;
the user behavior information class includes at least one or more of the following attributes: log in the location, history adds recommendation preferences.
Optionally, determining a first weight for each of the categories in each of the first user representation comprises:
for any first user representation;
obtaining initial weights of various types in any first user portrait and the latest updating date of the initial weights;
calculating the number of days between the current date and the latest update date of the initial weight of any first user portrait to obtain the number of days between any first user portrait;
first weights for each of the first user representations, initial weights for each of the first user representations, exp (-cooling coefficients for each of the first user representations, number of days between which the first user representation is spaced).
Optionally, the cooling coefficient of any of the first user representations is a coefficient of any/(number of times the user corresponding to the first user representation is recommended) number of attributes included in any of the first user representations);
the coefficient of the historical service information class is 1, the coefficient of the basic attribute class is 0.5, the coefficient of the company attribute class is 0.5, the coefficient of the interest class is 0.5, and the coefficient of the user behavior information class is 0.25.
Optionally, determining a second weight for each type in the second user representation comprises:
determining ideal weights of the user to be recommended for various types;
and the second weight of each type in the second user portrait is the initial weight of each type in the second user portrait, and the ideal weight/average weight of each type of the user to be recommended to each type.
Optionally, after recommending the user corresponding to the most similar first user figure to the user to be recommended, the method further includes:
updating various initial weights of each first user portrait to first weights;
updating various initial weights of the second user portrait to second weights;
a second user representation is stored.
Optionally, determining a first user representation that is most similar to the second user representation among all first user representations based on the first and second weights comprises:
for any first user portrait, sequentially selecting any first user portraitSemantic analysis is respectively carried out on each attribute value in the selected class and the corresponding attribute value of the corresponding class of the second user portrait, whether the semantics of the two classes are the same or not is determined, and the number of the same attributes of the selected class is obtained; calculating the similarity (Σ) of any one of the first user portraitsiexp (— first weight of ith class in any one of the first user representations-second weight of corresponding class in the second user representation) | first weight of ith class in any one of the first user representations) the number of the same attributes corresponding to each of the classes; wherein i is the identification of the class;
and displaying the first user portrait with the maximum similarity.
In a second aspect, embodiments of the present application provide a computer storage medium having a computer program stored thereon, where the computer program is executed by a processor to implement the steps of the method as described above.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory and one or more processors, where the memory is used to store one or more programs; the one or more programs, when executed by the one or more processors, implement the method as described above.
In the scheme provided in the embodiment, various first weights in each first user portrait and various second weights in each second user portrait are determined, similar users are determined according to the first user portrait, the first weights, the second user portrait and the second weights, and the weights are recalculated every time, so that the calculated weights are more in line with the current actual conditions of the stored users and the users to be recommended, similar user businesses obtained according to the weights are more similar to the users to be recommended, and the recommendation accuracy is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 shows a schematic flow chart of a recommendation method provided in an embodiment of the present application;
fig. 2 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application.
Detailed Description
Traditional offline social contact has few social opportunities and low efficiency. Many online logistics platforms also come up with some functionality for recommending users. But these functions generally query other users who have published compliant requirements for recommendations according to the requirements that the users published in their platforms. But such recommendations are less accurate.
Based on the above, the recommendation method is provided, various first weights in each first user portrait and various second weights in each second user portrait are determined, similar users are determined according to the first user portrait, the first weights, the second user portrait and the second weights, and the weights are recalculated each time, so that the calculated weights are more consistent with the current actual conditions of the stored users and the users to be recommended, the similar user businesses obtained according to the weights are more similar to the users to be recommended, and the recommendation accuracy is improved.
Referring to fig. 1, the implementation details of the recommendation method provided in this embodiment are as follows:
101, a first user image of each stored user and a second user image of a user to be recommended are obtained.
Wherein the user representation includes a classification.
The first user representation in this step is pre-stored. The storage process may collect user portrait related information on a regular basis (e.g., daily) and create a new user portrait archive for new users and update the portrait for older users.
The process of collecting the relevant information may be as follows: acquiring basic data records (such as account information, user company information, personal interest preference of a user and the like) of a user account, service data (main service operation records transported by the user in all subsystems of a platform), real-time positioning information (pc terminal id or mobile phone positioning data used by a client) and the like.
In addition, the classification of user representations includes (either the first user representation or the second user representation includes the following): basic attribute class, company attribute class, interest and hobby class, historical service information class and user behavior information class.
The basic property class includes at least one or more of the following properties: age, working age, location of mobile phone number.
The company attributes include at least one or more of the following: company business scope, company size, province of company business, company partner.
The interest taste class includes at least one or more of the following attributes: personal points of interest, industry categories of interest, business categories of interest, professional categories of interest.
The historical service information class includes at least one or more of the following attributes: service type, service occurrence time and service occurrence place.
The user behavior information class includes at least one or more of the following attributes: log in the location, history adds recommendation preferences.
In addition, the basic attribute class, the company attribute class, the interest and hobby class, the historical service information class and the user behavior information class all have initial weights and the latest updating dates of the initial weights.
The initial weight of the basic attribute class is 3, the initial weight of the company attribute class is 2, the initial weight of the interest and hobby class is 2, the initial weight of the historical service information class is 2, and the initial weight of the user behavior information class is 1.
The second user image is obtained according to the requirements of the user to be recommended, such as the age level, the location of the mobile phone, the interest point and the like of the user expected to be recommended by the user to be recommended. If the service type label is car finding, the corresponding expected label is that the company type is a motorcade or the service type is a carrier.
Here, 3, 2, 2, 2, 1 is only a specific gravity value, i.e., an initial weight of the basic attribute class: initial weight of company attribute class: initial weights of interest and hobby classes: initial weight of historical service information class: the initial weight of the user behavior information class is 3, 2, 2, 2, 1. The specific weight values may be 3, 2, 2, 2, 1, or 6, 4, 4, 4, 2, respectively. The present embodiment does not limit the specific values.
102, a first weight for each category in the first user representation and a second weight for each category in the second user representation are determined.
1) The determination of the first weights for each of the categories in each of the first user representation is accomplished as follows:
for any of the first user representation, the user is presented with a first user representation,
an initial weight for each type of first user representation and a most recent update date for the initial weight are obtained.
The number of days between the current date and the most recent update date of the initial weight of any first user representation is calculated to obtain the number of days between any first user representation.
The first weight of each category in any first user representation is the initial weight of each category in any first user representation, exp (cooling coefficient of each category in any first user representation, number of days between any first user representation).
The exp () is an exponential function with a natural constant e as a base, and the cooling coefficient of any one of the first user images is a coefficient of any one of the first user images/(the recommended times of the user corresponding to any one of the first user images is the number of attributes included in any one of the first user images).
The coefficient of the historical service information class is 1, the coefficient of the basic attribute class is 0.5, the coefficient of the company attribute class is 0.5, the coefficient of the interest class is 0.5, and the coefficient of the user behavior information class is 0.25.
The recommended times of the user corresponding to any one of the first user representations is the total recommended times of the user corresponding to any one of the first user representations in the history.
For example: for any first user portrait A, basic attribute classes, company attribute classes, interest classes and historical service information classes in the first user portrait A are determined, initial weights of the user behavior information classes are respectively 3, 2, 2, 2 and 1, and the latest update date of the initial weights is 2019, 6, 20.
If the current date is 2019, 6, month, 25, the number of days between which any of the first user images is obtained is 5. The company attribute class weight of the first user image a is 3 x exp (-cooling coefficient of the company attribute class in the first user image a 5).
If the company attribute class in the first user image a includes the company business range, the company scale, the province where the company business is located, the company partner, and 4 attributes, the cooling coefficient of the company attribute class in the first user image a is 0.5/(the recommended times of the user a corresponding to the first user image a is 4).
2) Determining the second weight for each type of the second user representation is accomplished as follows:
and determining the ideal weight of the user to be recommended to each type. And the second weight of each type in the second user portrait is the initial weight of each type in the second user portrait and the ideal weight/average weight of each type of the user to be recommended.
The ideal weights of the user to be recommended for various types are provided when the user inputs recommendation requirements, and the average weights of various types are the average weights of various types in the historical data.
103, a first user representation that is most similar to the second user representation is determined among all of the first user representations based on the first weight and the second weight.
For any first user portrait, one type of any first user portrait is selected in sequence, semantic analysis is carried out on each attribute value in the selected type and the corresponding attribute value of the corresponding type of the second user portrait respectively, whether the semantics of the two types are the same or not is determined, and the number of the same attributes of the selected type is obtained. Calculating the similarity (Σ) of any first user portraitiexp (| any first weight of ith class in the first user representation-second weight of corresponding class in the second user representation) | any first weight of ith class in the first user representation. Wherein i is the identification of the class.
And displaying the first user portrait with the maximum similarity.
For example, for a first user representation A, the age of the first user representation A (e.g., age 20) is semantically analyzed for the age of the base attribute class in the first user representation A, and for a second user representation (e.g., age 20) is semantically analyzed for the age of the base attribute class in the second user representation A. Semantic analysis is performed on the working age of the basic attribute class in the first user profile A to obtain the working age (such as 0 year) of the first user profile A, and semantic analysis is performed on the working age of the basic attribute class in the second user profile to obtain the working age (such as 0 year) of the second user profile, so that the semantic meanings of the two are the same. Semantic analysis is carried out on the mobile phone number of the basic attribute class in the first user portrait A to obtain the slave mobile phone number of the first user portrait A (such as Shanghai), and semantic analysis is carried out on the mobile phone number of the basic attribute class in the second user portrait to obtain the mobile phone number of the second user portrait (such as Beijing), and the semantics of the two mobile phone numbers are different. The number of attributes for which the basic attribute class is identical is 2.
A similarity of the first user representation a [ exp (| first weight of the base attribute class in the first user representation a-second weight of the base attribute class in the second user representation |) ] a first weight of the base attribute class in the first user representation a [ + number of same attributes of the base attribute class ] + [ exp (| first weight of the company attribute class in the first user representation a-second weight of the company attribute class in the second user representation |) ] a first weight of the company attribute class in the first user representation a [ + number of same attributes of the company attribute class ] + [ exp (| first weight of the interest class in the first user representation a-second weight of the interest class in the second user |) ] a first weight of the interest class in the first user representation a | number of same attributes of interest class ] + [ exp information of the first user representation a | first weight of the interest class in the first user representation a-second weight of the first user representation a | Second weight of historical service information class |). first weight of historical service information class in first user portrait A [ + [ exp (| first weight of user behavior information class in first user portrait A-second weight of user behavior information class in second user portrait |). first weight of user behavior information class in first user portrait A | ] same attribute quantity of user behavior information class in first user portrait A.
And 104, recommending the user corresponding to the most similar first user portrait to the user to be recommended.
In addition, in order to increase the self-learning capability of the recommendation method provided in this embodiment, after recommending the user corresponding to the most similar first user portrait to the user to be recommended, the weight calculated this time is also used as the initial weight for the next calculation, for example: the initial weights of the first user representations are updated to first weights. The initial weights of the various types of second user representations are updated to second weights.
In addition, the related image of the user to be recommended at this time is stored to be used as a candidate recommendation object when the user is recommended next time. Thus, a second user representation is also stored.
The recommendation algorithm provided by the embodiment is based on the whole set of logistics comprehensive platform, and is suitable for users across industries and functional posts in the field of logistics. The method can extract data in daily operation of a user, collect the data to form a user portrait and update the user portrait periodically. When the user triggers the recommendation function, the algorithm can match the portraits of all users according to the user portraits of the user, and finally a batch of recommendations which are most consistent are obtained and are recommended to the user.
The terms "first" and "second" in this example are merely labels, and are used to distinguish different user images, weights, etc., without any substantial meaning.
The method provided by the embodiment determines various first weights in each first user portrait and various second weights in each second user portrait, determines similar users according to the first user portrait, the first weights, the second user portrait and the second weights, and determines the weights of the similar users each time.
Based on the same inventive concept, the present embodiment provides a computer storage medium on which a computer program is stored, which when executed by a processor implements the following steps.
Acquiring a first user portrait of each stored user and a second user portrait of a user to be recommended; the user representation includes a classification;
determining a first weight of each type in each first user representation and a second weight of each type in the second user representation;
determining a first user representation that is most similar to the second user representation among all of the first user representations based on the first weight and the second weight;
and recommending the user corresponding to the most similar first user portrait to the user to be recommended.
Optionally, the classification of the user representation comprises: basic attribute class, company attribute class, interest and hobby class, historical service information class and user behavior information class;
the basic attribute class, the company attribute class, the interest and hobby class, the historical service information class and the user behavior information class all have initial weights and the latest updating dates of the initial weights;
the initial weight of the basic attribute class is 3, the initial weight of the company attribute class is 2, the initial weight of the interest and hobby class is 2, the initial weight of the historical service information class is 2, and the initial weight of the user behavior information class is 1.
Optionally, the basic property class includes at least one or more of the following properties: age, working age, location of mobile phone number;
the company attributes include at least one or more of the following: company business range, company scale, province of company business, company partner;
the interest taste class includes at least one or more of the following attributes: personal interest points, industry categories of interest, business categories of interest, professional categories of interest;
the historical service information class includes at least one or more of the following attributes: service type, service occurrence time and service occurrence place;
the user behavior information class includes at least one or more of the following attributes: log in the location, history adds recommendation preferences.
Optionally, determining a first weight for each of the categories in each of the first user representation comprises:
for any first user representation;
obtaining initial weights of various types in any first user portrait and the latest updating date of the initial weights;
calculating the number of days between the current date and the latest update date of the initial weight of any first user portrait to obtain the number of days between any first user portrait;
the first weight of each category in any first user representation is the initial weight of each category in any first user representation, exp (cooling coefficient of each category in any first user representation, number of days between any first user representation).
Optionally, the cooling coefficient of any one of the first user images is a coefficient of any one of the first user images/(the recommended number of times of the user corresponding to any one of the first user images is recommended) the number of attributes included in any one of the first user images);
the coefficient of the historical service information class is 1, the coefficient of the basic attribute class is 0.5, the coefficient of the company attribute class is 0.5, the coefficient of the interest class is 0.5, and the coefficient of the user behavior information class is 0.25.
Optionally, determining a second weight for each type in the second user representation comprises:
determining ideal weights of the user to be recommended for various types;
and the second weight of each type in the second user portrait is the initial weight of each type in the second user portrait and the ideal weight/average weight of each type of the user to be recommended.
Optionally, after recommending the user corresponding to the most similar first user portrait to the user to be recommended, the method further includes:
updating various initial weights of each first user portrait to first weights;
updating various initial weights of the second user portrait to second weights;
a second user representation is stored.
Optionally, determining a first user representation that is most similar to the second user representation among all of the first user representations based on the first weight and the second weight comprises:
for any first user portrait, one type of any first user portrait is selected in sequence, semantic analysis is carried out on each attribute value in the selected type and the corresponding attribute value of the corresponding type of the second user portrait respectively, whether the semantics of the two types are the same or not is determined, and the number of the same attributes of the selected type is obtained; calculating the similarity (Σ) of any first user portraitiexp (— any first weight of the ith class in the first user representation-second weight of the corresponding class in the second user representation) | any first weight of the ith class in the first user representation-the same attribute quantity corresponding to each class; wherein i is the identification of the class;
and displaying the first user portrait with the maximum similarity.
The computer program stored on the computer storage medium provided by this embodiment determines various first weights in each first user portrait and various second weights in each second user portrait, and determines similar users according to the first user portrait, the first weights, the second user portrait and the second weights.
Based on the same inventive concept, the present embodiment provides an electronic device, see fig. 2, comprising a memory 201, a processor 202, a bus 203, and a computer program stored on the memory 201 and executable on the processor 202, wherein the processor 202 implements the following steps when executing the program.
Acquiring a first user portrait of each stored user and a second user portrait of a user to be recommended; the user representation includes a classification;
determining a first weight of each type in each first user representation and a second weight of each type in the second user representation;
determining a first user representation that is most similar to the second user representation among all of the first user representations based on the first weight and the second weight;
and recommending the user corresponding to the most similar first user portrait to the user to be recommended.
Optionally, the classification of the user representation comprises: basic attribute class, company attribute class, interest and hobby class, historical service information class and user behavior information class;
the basic attribute class, the company attribute class, the interest and hobby class, the historical service information class and the user behavior information class all have initial weights and the latest updating dates of the initial weights;
the initial weight of the basic attribute class is 3, the initial weight of the company attribute class is 2, the initial weight of the interest and hobby class is 2, the initial weight of the historical service information class is 2, and the initial weight of the user behavior information class is 1.
Optionally, the basic property class includes at least one or more of the following properties: age, working age, location of mobile phone number;
the company attributes include at least one or more of the following: company business range, company scale, province of company business, company partner;
the interest taste class includes at least one or more of the following attributes: personal interest points, industry categories of interest, business categories of interest, professional categories of interest;
the historical service information class includes at least one or more of the following attributes: service type, service occurrence time and service occurrence place;
the user behavior information class includes at least one or more of the following attributes: log in the location, history adds recommendation preferences.
Optionally, determining a first weight for each of the categories in each of the first user representation comprises:
for any first user representation;
obtaining initial weights of various types in any first user portrait and the latest updating date of the initial weights;
calculating the number of days between the current date and the latest update date of the initial weight of any first user portrait to obtain the number of days between any first user portrait;
the first weight of each category in any first user representation is the initial weight of each category in any first user representation, exp (cooling coefficient of each category in any first user representation, number of days between any first user representation).
Optionally, the cooling coefficient of any one of the first user images is a coefficient of any one of the first user images/(the recommended number of times of the user corresponding to any one of the first user images is recommended) the number of attributes included in any one of the first user images);
the coefficient of the historical service information class is 1, the coefficient of the basic attribute class is 0.5, the coefficient of the company attribute class is 0.5, the coefficient of the interest class is 0.5, and the coefficient of the user behavior information class is 0.25.
Optionally, determining a second weight for each type in the second user representation comprises:
determining ideal weights of the user to be recommended for various types;
and the second weight of each type in the second user portrait is the initial weight of each type in the second user portrait and the ideal weight/average weight of each type of the user to be recommended.
Optionally, after recommending the user corresponding to the most similar first user portrait to the user to be recommended, the method further includes:
updating various initial weights of each first user portrait to first weights;
updating various initial weights of the second user portrait to second weights;
a second user representation is stored.
Optionally, determining a first user representation that is most similar to the second user representation among all of the first user representations based on the first weight and the second weight comprises:
for any first user portrait, one type of any first user portrait is selected in sequence, semantic analysis is carried out on each attribute value in the selected type and the corresponding attribute value of the corresponding type of the second user portrait respectively, whether the semantics of the two types are the same or not is determined, and the number of the same attributes of the selected type is obtained; calculating the similarity (Σ) of any first user portraitiexp (| any first weight of ith class in first user portrait-second weight |) of corresponding class in second user portrait |)The first weight of the ith class in the user portrait and the quantity of the same attributes corresponding to each class; wherein i is the identification of the class;
and displaying the first user portrait with the maximum similarity.
The electronic equipment provided by the embodiment determines various first weights in each first user portrait and various second weights in each second user portrait, determines similar users according to the first user portrait, the first weights, the second user portrait and the second weights, and determines the weights of the similar users each time.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
Claims (10)
1. A recommendation method, comprising:
acquiring a first user portrait of each stored user and a second user portrait of a user to be recommended; the user representation includes a classification;
determining a first weight of each type in each first user representation and a second weight of each type in the second user representation;
determining a first user representation that is most similar to the second user representation among all first user representations based on the first and second weights;
and recommending the user corresponding to the most similar first user portrait to the user to be recommended.
2. The method of claim 1, wherein the classification of the user representation comprises: basic attribute class, company attribute class, interest and hobby class, historical service information class and user behavior information class;
the basic attribute class, the company attribute class, the interest and hobby class, the historical service information class and the user behavior information class all have initial weights and the latest updating dates of the initial weights;
the initial weight of the basic attribute class is 3, the initial weight of the company attribute class is 2, the initial weight of the interest and hobby class is 2, the initial weight of the historical service information class is 2, and the initial weight of the user behavior information class is 1.
3. The method of claim 2, wherein the basic property class comprises at least one or more of the following properties: age, working age, location of mobile phone number;
the company attributes include at least one or more of the following attributes: company business range, company scale, province of company business, company partner;
the interest taste class includes at least one or more of the following attributes: personal interest points, industry categories of interest, business categories of interest, professional categories of interest;
the historical service information class includes at least one or more of the following attributes: service type, service occurrence time and service occurrence place;
the user behavior information class includes at least one or more of the following attributes: log in the location, history adds recommendation preferences.
4. The method of claim 2, wherein determining a first weight for each of the categories in each of the first user representation comprises:
for any first user representation;
obtaining initial weights of various types in any first user portrait and the latest updating date of the initial weights;
calculating the number of days between the current date and the latest update date of the initial weight of any first user portrait to obtain the number of days between any first user portrait;
first weights for each of the first user representations, initial weights for each of the first user representations, exp (-cooling coefficients for each of the first user representations, number of days between which the first user representation is spaced).
5. The method of claim 4, wherein the cooling coefficient for any of the first user representations is any coefficient/(number of times the user corresponding to the first user representation is recommended) number of attributes included in any of the first user representations);
the coefficient of the historical service information class is 1, the coefficient of the basic attribute class is 0.5, the coefficient of the company attribute class is 0.5, the coefficient of the interest class is 0.5, and the coefficient of the user behavior information class is 0.25.
6. The method of claim 2, wherein determining second weights for each type of second user representation comprises:
determining ideal weights of the user to be recommended for various types;
and the second weight of each type in the second user portrait is the initial weight of each type in the second user portrait, and the ideal weight/average weight of each type of the user to be recommended to each type.
7. The method of claim 2, wherein after recommending the user corresponding to the most similar first user representation to the user to be recommended, the method further comprises:
updating various initial weights of each first user portrait to first weights;
updating various initial weights of the second user portrait to second weights;
a second user representation is stored.
8. The method of claim 3, wherein determining a first user representation that is most similar to the second user representation among all first user representations based on the first and second weights comprises:
for any first user portrait, sequentially selecting one type of the first user portrait, performing semantic analysis on each attribute value in the selected type and the corresponding attribute value of the corresponding type of the second user portrait respectively, and determining whether the semantics of the two types are the same to obtain the number of the same attributes of the selected type; calculating the similarity (Sigma) of any first user portraitiexp (— first weight of ith class in any one of the first user representations-second weight of corresponding class in the second user representation) | first weight of ith class in any one of the first user representations) the number of the same attributes corresponding to each of the classes; wherein i is the identification of the class;
and displaying the first user portrait with the maximum similarity.
9. A computer storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 8.
10. An electronic device comprising one or more processors, and memory for storing one or more programs; the one or more programs, when executed by the one or more processors, implement the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910846332.4A CN110727858A (en) | 2019-09-09 | 2019-09-09 | Recommendation method, computer storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910846332.4A CN110727858A (en) | 2019-09-09 | 2019-09-09 | Recommendation method, computer storage medium and electronic device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110727858A true CN110727858A (en) | 2020-01-24 |
Family
ID=69217914
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910846332.4A Pending CN110727858A (en) | 2019-09-09 | 2019-09-09 | Recommendation method, computer storage medium and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110727858A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113672703A (en) * | 2021-08-26 | 2021-11-19 | 国家电网有限公司大数据中心 | User information updating method, device, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110145270A1 (en) * | 2009-12-14 | 2011-06-16 | Telefonaktiebolaget Lm Ericsson (Publ) | Service personas for address books |
| CN105608171A (en) * | 2015-12-22 | 2016-05-25 | 青岛海贝易通信息技术有限公司 | User portrait construction method |
| CN107145536A (en) * | 2017-04-19 | 2017-09-08 | 畅捷通信息技术股份有限公司 | User's portrait construction method and device and recommendation method and apparatus |
| CN109241425A (en) * | 2018-08-31 | 2019-01-18 | 腾讯科技（深圳）有限公司 | A kind of resource recommendation method, device, equipment and storage medium |
| CN109992982A (en) * | 2019-04-11 | 2019-07-09 | 北京信息科技大学 | Big data access authorization methods, device and big data platform |
| CN110046303A (en) * | 2019-04-09 | 2019-07-23 | 有光创新（北京）信息技术有限公司 | A kind of information recommendation method and device realized based on demand Matching Platform |
-
2019
- 2019-09-09 CN CN201910846332.4A patent/CN110727858A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110145270A1 (en) * | 2009-12-14 | 2011-06-16 | Telefonaktiebolaget Lm Ericsson (Publ) | Service personas for address books |
| CN105608171A (en) * | 2015-12-22 | 2016-05-25 | 青岛海贝易通信息技术有限公司 | User portrait construction method |
| CN107145536A (en) * | 2017-04-19 | 2017-09-08 | 畅捷通信息技术股份有限公司 | User's portrait construction method and device and recommendation method and apparatus |
| CN109241425A (en) * | 2018-08-31 | 2019-01-18 | 腾讯科技（深圳）有限公司 | A kind of resource recommendation method, device, equipment and storage medium |
| CN110046303A (en) * | 2019-04-09 | 2019-07-23 | 有光创新（北京）信息技术有限公司 | A kind of information recommendation method and device realized based on demand Matching Platform |
| CN109992982A (en) * | 2019-04-11 | 2019-07-09 | 北京信息科技大学 | Big data access authorization methods, device and big data platform |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113672703A (en) * | 2021-08-26 | 2021-11-19 | 国家电网有限公司大数据中心 | User information updating method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109033101B (en) | Label recommendation method and device | |
| CN111552870A (en) | Object recommendation method, electronic device and storage medium | |
| CN109345417B (en) | Online assessment method and terminal equipment for business personnel based on identity authentication | |
| CN115002200B (en) | Message pushing method, device, equipment and storage medium based on user portrait | |
| CN107808346B (en) | Evaluation method and evaluation device for potential target object | |
| EP3608799A1 (en) | Search method and apparatus, and non-temporary computer-readable storage medium | |
| CN113886691A (en) | Intelligent recommendation method and device based on historical data, electronic equipment and medium | |
| CN111814056A (en) | Supplier recommendation method based on information processing and related equipment | |
| CN117114514B (en) | Talent information analysis management method, system and device based on big data | |
| CN111913954A (en) | Intelligent data standard catalog generation method and device | |
| CN110059172B (en) | Method and device for recommending answers based on natural language understanding | |
| CN113326363A (en) | Searching method and device, prediction model training method and device, and electronic device | |
| CN110727858A (en) | Recommendation method, computer storage medium and electronic device | |
| CN111222923B (en) | Method and device for judging potential clients, electronic equipment and storage medium | |
| CN113327132A (en) | Multimedia recommendation method, device, equipment and storage medium | |
| CN116228301A (en) | Method, device, equipment and medium for determining target user | |
| CN115098766B (en) | Bidding information recommendation method and system for electronic bidding transaction platform | |
| CN110309406B (en) | Click rate estimation method, device, equipment and storage medium | |
| US20230237076A1 (en) | Automatically drawing infographics for statistical data based on a data model | |
| CN116644102A (en) | Intelligent investment object selection method, system terminal and computer readable storage medium | |
| CN112560433B (en) | Information processing method and device | |
| JP2020004161A (en) | Examination support apparatus, examination support method, and service providing method | |
| CN115186179A (en) | Insurance product pushing method and device | |
| CN113158037A (en) | Object-oriented information recommendation method and device | |
| CN111383050A (en) | Product data integration and analysis method, device and computer readable storage medium |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210319 Address after: 200333 room 3110, No. 100, Lane 130, Taopu Road, Putuo District, Shanghai Applicant after: Shanghai zebra Laila Logistics Technology Co.,Ltd. Address before: Room 308-1, area C, 1718 Daduhe Road, Putuo District, Shanghai 200333 Applicant before: Shanghai kjing XinDa science and Technology Group Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200124 |