CN109034941B - Product recommendation method and device, computer equipment and storage medium - Google Patents
Product recommendation method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109034941B CN109034941B CN201810607862.9A CN201810607862A CN109034941B CN 109034941 B CN109034941 B CN 109034941B CN 201810607862 A CN201810607862 A CN 201810607862A CN 109034941 B CN109034941 B CN 109034941B
- Authority
- CN
- China
- Prior art keywords
- product
- user
- data
- scoring matrix
- matrix
- 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.)
- Active
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
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a product recommendation method, a product recommendation device, computer equipment and a storage medium, wherein the method comprises the following steps: the method comprises the steps of obtaining user data of a certain user and product data of a certain product, obtaining a first scoring matrix of the user to the product according to the user data and obtaining a second scoring matrix of the product to the user according to the product data, wherein the user data are data of the certain user to all products, and the product data are data of the certain product to all users; and inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain user characteristics, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics. The product recommendation method, the product recommendation device, the computer equipment and the storage medium provided by the invention fully utilize all information of all users, accurately predict the interest degree of the users in different products and recommend reasonable products for the users.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a product recommendation method, apparatus, computer device, and storage medium.
Background
With the rapid development of society, a user recommendation system provides commodity information and suggestions to a user by using historical data information of the user, helps the user to decide what product should be purchased, and simulates a salesperson to help the customer complete a purchasing process. The personalized recommendation is to recommend information and commodities which are interested by the user to the user according to the interest characteristics and purchasing behaviors of the user. However, most of the existing user recommendation systems adopt implicit feedback as input, which has the disadvantage of causing information loss; the existing user recommendation system still faces the problem of data sparseness, so that the problem that all information of all users can be fully utilized, the interest degree of the users in different products is accurately predicted, and the recommendation of reasonable products for the users becomes urgent to solve is provided.
Disclosure of Invention
The invention mainly aims to provide a product recommendation method, a product recommendation device, computer equipment and a storage medium, which fully utilize all information of all users, accurately predict the interest degree of the users in different products and recommend reasonable products to the users.
The product recommendation method comprises the following steps:
the method comprises the steps of obtaining user data of a certain user and product data of a certain product, obtaining a first scoring matrix of the user for the product according to the user data and obtaining a second scoring matrix of the product for the user according to the product data, wherein the user data is data of the certain user for all products, and the product data is data of the certain product for all users;
inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain user characteristics, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics;
inputting the user characteristics and the product characteristics into a preset formula for calculation to obtain similarity values of the user characteristics and the product characteristics;
judging whether the similarity value is larger than a preset value or not;
and if so, recommending the product to the user.
Further, the method for generating the first scoring matrix includes:
and acquiring user data of a certain user, and setting corresponding scoring parameters according to preference degrees of the user on all products in multiple dimensions to construct and obtain the first scoring matrix.
Further, the method for generating the second scoring matrix includes:
and acquiring product data of a certain product, and setting corresponding grading parameter construction according to the applicability of the product to all users in multiple dimensions to obtain the second grading matrix.
Further, before the steps of inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain a user characteristic, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain a product characteristic, the method includes:
and respectively carrying out data expansion processing on the first scoring matrix and the second scoring matrix.
Further, the step of inputting the user characteristic and the product characteristic into a preset formula for calculation to obtain a similarity value between the user characteristic and the product characteristic includes:
according to the formula
Calculating to obtain the similarity value of the user characteristic and the product characteristic, wherein p i For said user features, q j Is characteristic of the product, is>
Is the similarity value.
Further, the step of recommending the product to the user includes:
matching the similarity value with a preset recommendation level table, wherein the recommendation level table comprises corresponding relations between different similarity value ranges and recommendation levels;
outputting a recommendation grade according to a matching result;
and recommending the product to the user according to the recommendation level.
Further, the method for respectively performing data expansion processing on the first scoring matrix and the second scoring matrix comprises:
and performing data expansion in a mode of extracting partial matrixes in the first scoring matrix and the second scoring matrix.
The product recommendation device provided by the invention comprises:
the system comprises a construction unit, a data acquisition unit and a data processing unit, wherein the construction unit is used for acquiring user data of a certain user and product data of a certain product, acquiring a first rating matrix of the user for the product according to the user data and a second rating matrix of the product for the user according to the product data, the user data is data of the certain user for all the products, and the product data is data of the certain product for all the users;
the first calculation unit is used for inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain user characteristics, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics;
the second calculation unit is used for inputting the user characteristics and the product characteristics into a preset formula for calculation to obtain similarity values of the user characteristics and the product characteristics;
the judging unit is used for judging whether the similarity value is larger than a preset value or not;
and the execution unit is used for recommending the product to the user when the similarity value is larger than a preset value.
The computer device provided by the invention comprises a memory and a processor, wherein the memory stores a computer program, and is characterized in that the processor implements the steps of the method when executing the computer program.
The present invention proposes a computer-readable storage medium having a computer program stored thereon, wherein the computer program is configured to implement the steps of the above method when executed by a processor.
The invention has the beneficial effects that: according to the similarity values of the user characteristics and the product characteristics obtained by the product recommendation method, the greater the similarity value is, the more interesting the user is to the product is indicated, and when the similarity value is greater than a preset value, the product is recommended to the user, so that the interest degree of the user to the product is accurately predicted by fully utilizing various information and product characteristic information of the user, a reasonable product is recommended to the user, and the user experience is improved.
Drawings
FIG. 1 is a schematic diagram illustrating steps of a product recommendation method according to an embodiment of the invention;
FIG. 2 is a schematic diagram of the steps of a product recommendation method according to another embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a product recommendation device according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a product recommendation device in another embodiment of the present invention;
FIG. 5 is a schematic structural diagram of an execution unit of the product recommendation device according to an embodiment of the present invention;
fig. 6 is a block diagram schematically illustrating a structure of a computer apparatus according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, the product recommendation method in this embodiment includes:
the method comprises the following steps of S1, obtaining user data of a certain user and product data of a certain product, obtaining a first rating matrix of the user for the product according to the user data and obtaining a second rating matrix of the product for the user according to the product data, wherein the user data is the data of the certain user for all the products, and the product data is the data of the certain product for all the users;
s2, inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain user characteristics, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics;
s3, inputting the user characteristics and the product characteristics into a preset formula for calculation to obtain similarity values of the user characteristics and the product characteristics;
s4, judging whether the similarity value is larger than a preset value or not;
and S5, if yes, recommending the product to the user.
In step S1, user data of a certain user and product data of a certain product are obtained, where the products are insurance products of different types of insurance companies, and the product data of the certain product includes various product characteristic information such as income conditions, applicable ages, applicable professions, and the like, applicable to all users by the product; the user data includes various information such as the user's evaluation of all products after purchase, the number of purchased products, the number of times the products are viewed, and the time at which the products are viewed. And for the user data and the product data, setting corresponding grading parameter structures according to the evaluation of the user on all products after the products are purchased, the number of purchased products, the frequency of browsing the products and the preference degree of the time for browsing the products in multiple dimensions to obtain a first grading matrix, and setting corresponding grading parameter structures according to the income condition, the applicable age and the applicable degree of the products in multiple dimensions to all users to obtain a second grading matrix.
In step S2, the first scoring matrix is input into a preset first depth matrix decomposition model for calculation to obtain a user characteristic, where the user characteristic is a matrix. The preset first depth matrix decomposition model needs to be trained, the preset first depth matrix decomposition model is trained by using a first scoring matrix with a specified amount and user features corresponding to the first scoring matrix as sample data, and the first depth matrix decomposition model is used for calculating the user features. And inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics, wherein the product characteristics are also matrixes. The preset second depth matrix decomposition model also needs to be trained, the preset second depth matrix decomposition model is trained by using a second scoring matrix with a specified amount and product features corresponding to the second scoring matrix as sample data, and the second depth matrix decomposition model is used for calculating the product features.
In step S3, the user characteristics and the product characteristics are input into a preset formula for calculation to obtain a calculation result, where the preset formula requires to input the user characteristics and the product characteristics at the same time, and the preset formula is
p i For the above-mentioned user characteristics, q j Is characterized by the above-mentioned product>
Calculating a result of ≥ for the similarity value between the user characteristic and the product characteristic>
I.e. the similarity values of the user characteristic and the product characteristic.
In step S4, for the obtained similarity values of the user characteristic and the product characteristic, when the similarity value is larger, it indicates that the user is more interested in the product. Therefore, whether the product needs to be recommended to the user can be judged according to the size of the similarity value, and therefore a corresponding preset value can be set to judge whether the similarity value is larger than the preset value.
In step S5, when the similarity value is greater than the preset value, the product is recommended to the user, so that various information and product characteristic information of the user are fully utilized, the interest degree of the user in the product is accurately predicted, a reasonable product is recommended to the user, and the user experience is improved.
In the product recommendation method in this embodiment, the generation method of the first scoring matrix includes:
and acquiring user data of a certain user, and setting corresponding scoring parameters according to preference degrees of the user on all products in multiple dimensions to construct and obtain the first scoring matrix.
The generation mode of the first scoring matrix specifically comprises the steps of firstly acquiring user data of a certain user, and setting corresponding scoring parameters according to preference degrees of multiple dimensions such as evaluation of all products of the user after purchase, the number of purchased products, the frequency of browsing the products, the time of browsing the products and the like to obtain the first scoring matrix. For example, the evaluation of the user after purchasing a certain product may be subdivided into evaluations of different grades such as a very good, a like, a general like, a dislike, a very dislike, and for the evaluations of different grades of a certain product by the user, the corresponding scoring parameters may be sequentially set to different scoring parameters such as 5, 4, 3, 2, and 1, so as to realize setting of the scoring parameters according to the evaluation of the certain product by the user. For the number of purchased products, the corresponding scoring parameters can be sequentially set to different scoring parameters such as 5, 4, 3, 2, 1 and the like according to different purchase times such as four times of purchase, three times of purchase, two times of purchase, one time of purchase, no purchase and the like, so that the scoring parameters can be set according to the number of purchased products. Similarly, for the information such as the number of times of browsing the product and the time of browsing the product, the above method can be adopted to set different scoring parameters, which is not described herein again. And setting the scoring parameters of the user to a certain product under the multiple dimensions to obtain a scoring vector, and constructing the scoring parameters of the user to all the products to obtain the first scoring matrix.
In the product recommendation method in this embodiment, the generation method of the second scoring matrix includes:
and acquiring product data of a certain product, and setting corresponding scoring parameter construction according to the applicability of the product to all users in multiple dimensions to obtain the second scoring matrix.
The generation method of the second scoring matrix specifically includes that product data of a certain product is obtained first, and corresponding scoring parameters are set according to the applicable degrees of the product to multiple dimensions such as income conditions, applicable ages and applicable professions of all users to construct the second scoring matrix. The selling price of a certain product is different, the users with the applicable income conditions are different, for example, the price of a certain product is very high, the corresponding scoring parameters are different for the users with different consumption levels, the scoring parameter of the product to the user is set to be 1 for the user with strong consumption capability, and the scoring parameter of the product to the user is set to be 0 for the user with weak consumption capability. The applicable age of a certain product is 20 to 30 years, the user in the age range of 20 to 30 years is set with the rating parameter of the product to the user as 1, and the user not in the age range of 20 to 30 years is set with the rating parameter of the product to the user as 0. Similarly, for information such as applicable occupation, different scoring parameters can be set by the method, which is not described herein again. And setting the scoring parameters of the product to a certain user under the multiple dimensions to obtain a scoring vector, and constructing the scoring parameters of the product to all the users to obtain the second scoring matrix.
Referring to fig. 2, before step S2 of inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain a user characteristic and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain a product characteristic, the product recommendation method in another embodiment includes:
For the first scoring matrix and the second scoring matrix, before the first scoring matrix is input into a preset first depth matrix decomposition model for calculation and the second scoring matrix is input into a preset second depth matrix decomposition model for calculation, data expansion processing can be respectively carried out on the first scoring matrix and the second scoring matrix, the size of the matrixes is expanded, the size of a training set is increased, and the purpose of avoiding the occurrence of overfitting is realized, so that the prediction accuracy of the first depth matrix decomposition model and the second depth matrix decomposition model is improved. For example, for the first scoring matrix, the specific way of performing data expansion processing is to extract a part of the matrix from the first scoring matrix as an expanded first expansion matrix, and then input the first expansion matrix and the first scoring matrix into the first depth matrix decomposition model together for calculation; the method for performing data expansion processing on the second scoring matrix is the same as the above method, and is not described herein again.
In the product recommendation method in this embodiment, the step S3 of inputting the user characteristic and the product characteristic into a preset formula to calculate to obtain a similarity value between the user characteristic and the product characteristic includes:
according to the formula
Calculating to obtain the similarity value of the user characteristic and the product characteristic, wherein p i For said user features, q j Is characteristic of the product, is>
Is the similarity value.
For the input user characteristics and product characteristics, calculating according to a preset formula to obtain similarity values of the user characteristics and the product characteristics, wherein the preset formula is
Wherein p is i For the above-mentioned user characteristics, q j Is characterized by the above-mentioned product>
And the similarity value is the similarity value of the user characteristic and the product characteristic and is used for representing the interest degree of the user in the product, and when the similarity value is larger, the interest degree of the user in the product is higher.
In the product recommendation method in this embodiment, the step S5 of recommending the product to the user includes:
step S51, matching the similarity value with a preset recommendation level table, wherein the recommendation level table comprises corresponding relations between different similarity value ranges and recommendation levels;
step S52, outputting a recommendation grade according to a matching result;
and step S53, recommending the product to the user according to the recommendation level.
For the output similarity values of the user characteristics and the product characteristics, matching the similarity value of the user interested in the product with a preset recommendation level table, wherein the recommendation level table comprises corresponding relations between different similarity value ranges and recommendation levels
The larger the value of (a), the higher the degree of interest of the user in the product; outputting corresponding recommendation levels according to the matching result, wherein the recommendation can be different recommendation levels such as strong recommendation and recommendation, and the like, and the recommendation levels are set according to the recommendation levelsThe user recommends the product, and reasonable products are recommended to the user in a more humanized mode.
Referring to fig. 3, the product recommendation apparatus in this embodiment includes:
the system comprises a construction unit 10, a data processing unit and a data processing unit, wherein the construction unit 10 is used for acquiring user data of a certain user and product data of a certain product, acquiring a first scoring matrix of the user for the product according to the user data and acquiring a second scoring matrix of the product for the user according to the product data, the user data is data of the certain user for all products, and the product data is data of the certain product for all users;
the first calculating unit 20 is configured to input the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain a user characteristic, and input the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain a product characteristic;
the second calculating unit 30 is configured to input the user characteristic and the product characteristic into a preset formula to calculate, so as to obtain a similarity value between the user characteristic and the product characteristic;
a judging unit 40, configured to judge whether the similarity value is greater than a preset value;
and the execution unit 50 is configured to recommend the product to the user when the similarity value is greater than a preset value.
Acquiring user data of a certain user and product data of a certain product, wherein the product is insurance products of different types of insurance companies, and the product data of the certain product comprises income condition, applicable age, applicable occupation and other product characteristic information of the product applicable to all users; the user data includes various information such as the user's evaluation of all products after purchase, the number of purchased products, the number of times the products are viewed, and the time at which the products are viewed. For the user data and the product data, the construction unit 10 sets a corresponding scoring parameter construction according to the preference degrees of the user to multiple dimensions of the product purchase evaluation, the product purchase quantity, the product browsing times and the product browsing time of all products to obtain a first scoring matrix, and sets a corresponding scoring parameter construction according to the income condition, the applicable age and the applicable occupation applicability degrees of the products to all users to obtain a second scoring matrix.
The first calculating unit 20 inputs the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain a user characteristic, where the user characteristic is a matrix. The preset first depth matrix decomposition model needs to be trained, the preset first depth matrix decomposition model is trained by using a first scoring matrix with a specified amount and user features corresponding to the first scoring matrix as sample data, and the first depth matrix decomposition model is used for calculating the user features. And inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics, wherein the product characteristics are also matrixes. The preset second depth matrix decomposition model also needs to be trained, the preset second depth matrix decomposition model is trained by using a second scoring matrix with a specified amount and the product features corresponding to the first person scoring matrix as sample data, and the second depth matrix decomposition model is used for calculating the product features.
The second calculating unit 30 inputs the user characteristics and the product characteristics into a preset formula for calculation, wherein the preset formula requires the user characteristics and the product characteristics to be input simultaneously, and the preset formula is
p i For the above-mentioned user characteristics, q j Is characterized by the above-mentioned product>
Calculating a result of ≥ for the similarity value between the user characteristic and the product characteristic>
I.e. the similarity of the user characteristic and the product characteristicThe value is obtained.
And for the obtained similarity values of the user characteristics and the product characteristics, when the similarity value is larger, the user is indicated to be more interested in the product. Therefore, it can be determined whether the product needs to be recommended to the user according to the size of the similarity value, so that a corresponding preset value can be set, and the determining unit 40 determines whether the similarity value is greater than the preset value.
When the similarity value is greater than the preset value, the execution unit 50 recommends the product to the user, so that various information and product characteristic information of the user are fully utilized, the interest degree of the user in the product is accurately predicted, a reasonable product is recommended for the user, and the user experience is improved.
In the product recommendation device in this embodiment, the method for generating the first scoring matrix includes:
and acquiring user data of a certain user, and setting corresponding scoring parameters according to preference degrees of the user on all products in multiple dimensions to construct and obtain the first scoring matrix.
The generation mode of the first scoring matrix specifically comprises the steps of firstly acquiring user data of a certain user, and setting corresponding scoring parameters according to preference degrees of multiple dimensions such as evaluation of all products of the user after purchase, the number of purchased products, the frequency of browsing the products, the time of browsing the products and the like to obtain the first scoring matrix. For example, the evaluation of a user after purchasing a certain product may be subdivided into evaluations of different levels such as a high level, a low level, and a high level, and for the evaluations of different levels of a certain product by the user, the corresponding scoring parameters may be sequentially set to different scoring parameters such as 5, 4, 3, 2, and 1, so as to set the scoring parameters according to the evaluation of the certain product by the user. For the number of purchased products, the corresponding scoring parameters can be sequentially set to different scoring parameters such as 5, 4, 3, 2, 1 and the like according to different purchase times such as four times of purchase, three times of purchase, two times of purchase, one time of purchase, no purchase and the like, so that the scoring parameters can be set according to the number of purchased products. Similarly, for the information such as the number of times of browsing the product and the time of browsing the product, the above method can be adopted to set different scoring parameters, which is not described herein again. And setting the scoring parameters of the user to a certain product under the multiple dimensions to obtain a scoring vector, and constructing the scoring parameters of the user to all the products to obtain the first scoring matrix.
In the product recommendation device in this embodiment, the method for generating the second scoring matrix includes:
and acquiring product data of a certain product, and setting corresponding grading parameter construction according to preference degrees of the product to all users in multiple dimensions to obtain the second grading matrix.
The generation mode of the second scoring matrix specifically includes that product data of a certain product is obtained first, and corresponding scoring parameters are set according to the applicable degrees of the product to multiple dimensions such as income conditions, applicable ages and applicable occupations of all users to construct the second scoring matrix. The selling price of a certain product is different, the users of the applicable income condition are different, for example, the price of a certain product is very high, the corresponding scoring parameters are different for the users with different consumption levels, the scoring parameter of the product to the user is set to be 1 for the user with strong consumption capability, and the scoring parameter of the product to the user is set to be 0 for the user with weak consumption capability. The applicable age of a certain product is 20 years to 30 years, the user in the age range of 20 years to 30 years is set to have the rating parameter of the product to the user set to 1, and the user not in the age range of 20 years to 30 years is set to have the rating parameter of the product to the user set to 0. Similarly, for information suitable for career and the like, different scoring parameters can be set by adopting the method, which is not described herein again. And setting the scoring parameters of the product to a certain user under the multiple dimensions to obtain a scoring vector, and constructing the scoring parameters of the product to all the users to obtain the second scoring matrix.
Referring to fig. 4, the product recommendation device in another embodiment further includes:
an expansion unit 201, configured to perform data expansion processing on the first scoring matrix and the second scoring matrix respectively.
For the first scoring matrix and the second scoring matrix, before the first scoring matrix is input into a preset first depth matrix decomposition model for calculation and the second scoring matrix is input into a preset second depth matrix decomposition model for calculation, the expansion unit 201 performs data expansion processing on the first scoring matrix and the second scoring matrix respectively, expands the size of the matrices, and increases the size of a training set, and aims to avoid over-fitting and improve the accuracy of prediction of the first depth matrix decomposition model and the second depth matrix decomposition model. For example, for the first scoring matrix, the specific way of performing data expansion processing is to extract a part of matrix in the first scoring matrix as an expanded first expansion matrix, and then input the first expansion matrix and the first scoring matrix into the first depth matrix decomposition model together for calculation; the method for performing data expansion processing on the second scoring matrix is the same as the above method, and is not described herein again.
In the product recommendation apparatus in this embodiment, the second calculating unit 30 is specifically configured to calculate the product recommendation according to a formula
Calculating to obtain the similarity value of the user characteristic and the product characteristic, wherein p i For said user features, q j Is characteristic of the product, is>
Is the similarity value.
For the input user characteristics and product characteristics, calculating according to a preset formula to obtain similarity values of the user characteristics and the product characteristics, wherein the preset formula is
Wherein p is i For the above-mentioned user characteristics, q j Is characterized by the above-mentioned product>
And the similarity value is the similarity value of the user characteristic and the product characteristic and is used for representing the degree of the product interest of the user, and when the similarity value is larger, the degree of the product interest of the user is higher.
Referring to fig. 5, in the product recommendation apparatus in this embodiment, the execution unit 50 includes:
a first matching module 51, configured to match the similarity value with a preset recommendation level table, where the recommendation level table includes correspondence between different similarity value ranges and recommendation levels;
a second matching module 52, configured to output a recommendation level according to a matching result;
and the recommending module 53 is configured to recommend the product to the user according to the recommendation level.
For the similarity values of the output user characteristics and the output product characteristics, the first matching module 51 may further match the similarity values of the user's interest in the product with a preset recommendation level table, where the similarity values are obtained by comparing the user's interest in the product with the preset recommendation level table
The recommendation level table comprises corresponding relations between different similarity value ranges and recommendation levels
The larger the value of (a), the higher the degree of user interest in the product; the second matching module 52 outputs corresponding recommendation levels according to the matching result, wherein the recommendation levels can be different recommendation levels such as strong recommendation and recommendation, and the recommendation module 53 recommends the product to the user according to the recommendation levels, so that more humanized and reasonable product recommendation for the user is realized.
Referring to fig. 6, an embodiment of the present invention further provides a computer device, where the computer device may be a server, and an internal structure of the computer device may be as shown in fig. 6. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used for presetting data such as a product recommendation method and the like. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a product recommendation method.
The processor executes the steps of the product recommendation method: the method comprises the steps of obtaining user data of a certain user and product data of a certain product, obtaining a first scoring matrix of the user for the product according to the user data and obtaining a second scoring matrix of the product for the user according to the product data, wherein the user data is data of the certain user for all products, and the product data is data of the certain product for all users; inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain user characteristics, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics; inputting the user characteristics and the product characteristics into a preset formula for calculation to obtain similarity values of the user characteristics and the product characteristics; judging whether the similarity value is larger than a preset value or not; and if so, recommending the product to the user.
The computer device and the method for generating the first scoring matrix include: and acquiring user data of a certain user, and setting corresponding scoring parameters according to the preference degree of the user to all products in multiple dimensions to construct the first scoring matrix.
In an embodiment, the method for generating the second scoring matrix includes: and acquiring product data of a certain product, and setting corresponding grading parameter construction according to the applicability of the product to all users in multiple dimensions to obtain the second grading matrix.
In an embodiment, before the step of inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain a user characteristic and the step of inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain a product characteristic, the method includes: and respectively carrying out data expansion processing on the first scoring matrix and the second scoring matrix.
In an embodiment, the step of inputting the user characteristic and the product characteristic into a preset formula to calculate a similarity value between the user characteristic and the product characteristic includes: according to the formula
Calculating to obtain the similarity value of the user characteristic and the product characteristic, wherein p i For the above-mentioned user characteristics, q j In order to achieve the above-mentioned product characteristics,
the above similarity values.
In one embodiment, the step of recommending the product to the user includes: matching the similarity value with a preset recommendation level table, wherein the recommendation level table comprises corresponding relations between different similarity value ranges and recommendation levels; outputting a recommendation grade according to a matching result; and recommending the product to the user according to the recommendation level.
In an embodiment, the method for performing data expansion processing on the first scoring matrix and the second scoring matrix respectively includes: and performing data expansion in a mode of extracting partial matrixes in the first scoring matrix and the second scoring matrix.
Those skilled in the art will appreciate that the architecture shown in fig. 6 is merely a block diagram of some of the structures associated with the present teachings and is not intended to limit the scope of the present teachings as applied to computer devices.
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a product recommendation method, and specifically: the method comprises the steps of obtaining user data of a certain user and product data of a certain product, obtaining a first scoring matrix of the user to the product according to the user data and obtaining a second scoring matrix of the product to the user according to the product data, wherein the user data are data of the certain user to all products, and the product data are data of the certain product to all users; inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain user characteristics, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics; inputting the user characteristics and the product characteristics into a preset formula for calculation to obtain similarity values of the user characteristics and the product characteristics; judging whether the similarity value is larger than a preset value or not; and if so, recommending the product to the user.
The computer-readable storage medium is a method for generating the first scoring matrix, and includes: and acquiring user data of a certain user, and setting corresponding scoring parameters according to preference degrees of the user on all products in multiple dimensions to construct and obtain the first scoring matrix.
In an embodiment, the method for generating the second scoring matrix includes: and acquiring product data of a certain product, and setting corresponding grading parameter construction according to the applicability of the product to all users in multiple dimensions to obtain the second grading matrix.
In an embodiment, before the step of inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain a user characteristic and the step of inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain a product characteristic, the method includes: and respectively carrying out data expansion processing on the first scoring matrix and the second scoring matrix.
In an embodiment, the step of inputting the user characteristic and the product characteristic into a preset formula to calculate a similarity value between the user characteristic and the product characteristic includes: according to the formula
Calculating to obtain the similarity value of the user characteristic and the product characteristic, wherein p i For the above-mentioned user characteristics, q j In order to achieve the above-mentioned product characteristics,
is the above similarity value.
In one embodiment, the step of recommending the product to the user includes: matching the similarity value with a preset recommendation level table, wherein the recommendation level table comprises corresponding relations between different similarity value ranges and recommendation levels; outputting a recommendation grade according to a matching result; and recommending the product to the user according to the recommendation level.
In an embodiment, the method for performing data expansion processing on the first scoring matrix and the second scoring matrix respectively includes: and performing data expansion in a mode of extracting partial matrixes in the first scoring matrix and the second scoring matrix.
It will be understood by those skilled in the art that all or part of the processes of the methods of the above embodiments may be implemented by hardware associated with instructions of a computer program, which may be stored on a non-volatile computer-readable storage medium, and when executed, may include processes of the above embodiments of the methods. Any reference to memory, storage, database, or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (SSRDRAM), enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM).
In summary, according to the similarity values of the user features and the product features obtained by the product recommendation method in the present invention, when the similarity value is larger, the user is more interested in the product, and when the similarity value is larger than a preset value, the product is recommended to the user, so that the product interest degree of the user in the product is accurately predicted by fully utilizing various information and product characteristic information of the user, a reasonable product is recommended to the user, and the user experience is improved.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all equivalent structures or equivalent processes performed by the present invention or directly or indirectly applied to other related technical fields will be covered by the scope of the present invention.
Claims (8)
1. A method for recommending products, comprising:
the method comprises the steps of obtaining user data of a certain user and product data of a certain product, obtaining a first scoring matrix of the user for the product according to the user data and obtaining a second scoring matrix of the product for the user according to the product data, wherein the user data is data of the certain user for all products, and the product data is data of the certain product for all users;
inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain user characteristics, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics;
inputting the user characteristics and the product characteristics into a preset formula for calculation to obtain similarity values of the user characteristics and the product characteristics;
judging whether the similarity value is larger than a preset value or not;
if so, recommending the product to the user;
the generation method of the first scoring matrix comprises the following steps:
acquiring user data of a certain user, and setting corresponding scoring parameters according to preference degrees of the user on all products in multiple dimensions to construct a first scoring matrix;
the generation method of the second scoring matrix comprises the following steps:
and acquiring product data of a certain product, and setting corresponding grading parameter construction according to the applicability of the product to all users in multiple dimensions to obtain the second grading matrix.
2. The product recommendation method according to claim 1, wherein the step of inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain the user characteristics and the step of inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain the product characteristics is preceded by the steps of:
and respectively carrying out data expansion processing on the first scoring matrix and the second scoring matrix.
3. The product recommendation method according to claim 1, wherein the step of inputting the user characteristics and the product characteristics into a preset formula for calculation to obtain the similarity values of the user characteristics and the product characteristics comprises:
according to the formula
Calculating to obtain the similarity value of the user characteristic and the product characteristic, wherein p i For said user features, q j For said product characteristic, based on the number of product groups in the product group>
Is the similarity value.
4. The product recommendation method of claim 1, wherein the step of recommending the product to the user comprises:
matching the similarity value with a preset recommendation level table, wherein the recommendation level table comprises corresponding relations between different similarity value ranges and recommendation levels;
outputting a recommendation grade according to a matching result;
and recommending the product to the user according to the recommendation level.
5. The product recommendation method according to claim 2, wherein the method for performing data expansion processing on the first scoring matrix and the second scoring matrix respectively comprises:
and performing data expansion in a mode of extracting partial matrixes in the first scoring matrix and the second scoring matrix.
6. A product recommendation device, comprising:
the system comprises a construction unit, a data acquisition unit and a data processing unit, wherein the construction unit is used for acquiring user data of a certain user and product data of a certain product, acquiring a first rating matrix of the user for the product according to the user data and a second rating matrix of the product for the user according to the product data, the user data is data of the certain user for all the products, and the product data is data of the certain product for all the users;
the first calculation unit is used for inputting the first scoring matrix into a preset first depth matrix decomposition model for calculation to obtain user characteristics, and inputting the second scoring matrix into a preset second depth matrix decomposition model for calculation to obtain product characteristics;
the second calculation unit is used for inputting the user characteristics and the product characteristics into a preset formula for calculation to obtain similarity values of the user characteristics and the product characteristics;
the judging unit is used for judging whether the similarity value is larger than a preset value or not;
the execution unit is used for recommending the product to the user when the similarity value is larger than a preset value;
the generation method of the first scoring matrix comprises the following steps:
acquiring user data of a certain user, and setting corresponding scoring parameters according to preference degrees of the user on all products in multiple dimensions to construct a first scoring matrix;
the generation method of the second scoring matrix comprises the following steps:
and acquiring product data of a certain product, and setting corresponding grading parameter construction according to the applicability of the product to all users in multiple dimensions to obtain the second grading matrix.
7. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 5 when executing the computer program.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 5.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810607862.9A CN109034941B (en) | 2018-06-13 | 2018-06-13 | Product recommendation method and device, computer equipment and storage medium |
| PCT/CN2018/108033 WO2019237569A1 (en) | 2018-06-13 | 2018-09-27 | Product recommendation method and apparatus, and computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810607862.9A CN109034941B (en) | 2018-06-13 | 2018-06-13 | Product recommendation method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109034941A CN109034941A (en) | 2018-12-18 |
| CN109034941B true CN109034941B (en) | 2023-03-31 |
Family
ID=64609539
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810607862.9A Active CN109034941B (en) | 2018-06-13 | 2018-06-13 | Product recommendation method and device, computer equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109034941B (en) |
| WO (1) | WO2019237569A1 (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110083688B (en) * | 2019-05-10 | 2022-03-25 | 北京百度网讯科技有限公司 | Search result recall method, device, server and storage medium |
| CN110598040B (en) * | 2019-09-17 | 2023-10-10 | 上海喜马拉雅科技有限公司 | Album recall method, device, equipment and storage medium |
| CN113344673A (en) * | 2021-06-28 | 2021-09-03 | 平安信托有限责任公司 | Product pushing method and device, electronic equipment and storage medium |
| CN115880076A (en) * | 2022-04-20 | 2023-03-31 | 北京中关村科金技术有限公司 | Trusted product recommendation method, device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102426686A (en) * | 2011-09-29 | 2012-04-25 | 南京大学 | Internet information product recommending method based on matrix decomposition |
| CN103246672A (en) * | 2012-02-09 | 2013-08-14 | 中国科学技术大学 | Method and device for performing personalized recommendation on users |
| CN106570008A (en) * | 2015-10-09 | 2017-04-19 | 阿里巴巴集团控股有限公司 | Recommendation method and device |
| CN107563841A (en) * | 2017-08-03 | 2018-01-09 | 电子科技大学 | A kind of commending system decomposed that scored based on user |
| CN107578270A (en) * | 2017-08-03 | 2018-01-12 | ***股份有限公司 | A kind of construction method, device and the computing device of financial label |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663627A (en) * | 2012-04-26 | 2012-09-12 | 焦点科技股份有限公司 | Personalized recommendation method |
| CN104021163B (en) * | 2014-05-28 | 2017-10-24 | 深圳市盛讯达科技股份有限公司 | Products Show system and method |
| CN107292648A (en) * | 2016-04-01 | 2017-10-24 | 株式会社理光 | A kind of user behavior analysis method and device |
| CN107562795A (en) * | 2017-08-01 | 2018-01-09 | 广州市香港科大***研究院 | Recommendation method and device based on Heterogeneous Information network |
-
2018
- 2018-06-13 CN CN201810607862.9A patent/CN109034941B/en active Active
- 2018-09-27 WO PCT/CN2018/108033 patent/WO2019237569A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102426686A (en) * | 2011-09-29 | 2012-04-25 | 南京大学 | Internet information product recommending method based on matrix decomposition |
| CN103246672A (en) * | 2012-02-09 | 2013-08-14 | 中国科学技术大学 | Method and device for performing personalized recommendation on users |
| CN106570008A (en) * | 2015-10-09 | 2017-04-19 | 阿里巴巴集团控股有限公司 | Recommendation method and device |
| CN107563841A (en) * | 2017-08-03 | 2018-01-09 | 电子科技大学 | A kind of commending system decomposed that scored based on user |
| CN107578270A (en) * | 2017-08-03 | 2018-01-12 | ***股份有限公司 | A kind of construction method, device and the computing device of financial label |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109034941A (en) | 2018-12-18 |
| WO2019237569A1 (en) | 2019-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109034941B (en) | Product recommendation method and device, computer equipment and storage medium | |
| CN106251174A (en) | Information recommendation method and device | |
| CN111798273A (en) | Training method of purchase probability prediction model of product and purchase probability prediction method | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN111080398A (en) | Commodity recommendation method and device, computer equipment and storage medium | |
| CN112182384B (en) | Content recommendation method and device based on countermeasure learning and computer equipment | |
| CN108763305A (en) | Method, apparatus, computer equipment and the storage medium of data characteristics extraction | |
| KR101635283B1 (en) | Method for analyzing data based on matrix factorization model and apparatus therefor | |
| CN109801101A (en) | Label determines method, apparatus, computer equipment and storage medium | |
| CN113343091A (en) | Industrial and enterprise oriented science and technology service recommendation calculation method, medium and program | |
| CN114398059A (en) | Parameter updating method, device, equipment and storage medium | |
| CN112464660B (en) | Text classification model construction method and text data processing method | |
| CN112734307A (en) | Online real-time enterprise portrait generation method, device, equipment and storage medium | |
| CN111552810B (en) | Entity extraction and classification method, entity extraction and classification device, computer equipment and storage medium | |
| CN111489196B (en) | Prediction method and device based on deep learning network, electronic equipment and medium | |
| CN111161009A (en) | Information pushing method and device, computer equipment and storage medium | |
| CN108986786B (en) | Voice interaction equipment rating method, system, computer equipment and storage medium | |
| CN116955797A (en) | Resource recommendation method and device, electronic equipment and storage medium | |
| CN110766465A (en) | Financial product evaluation method and verification method and device thereof | |
| CN107977794B (en) | Data processing method and device for industrial product, computer equipment and storage medium | |
| CN113420876B (en) | Method, device and equipment for processing real-time operation data based on unsupervised learning | |
| EP3782079A1 (en) | Model interpretation | |
| CN114820113A (en) | E-commerce platform recommendation adjustment method and system based on block chain | |
| CN112989022B (en) | Intelligent virtual text selection method and device and computer equipment | |
| Belomestny et al. | Optimal stopping via reinforced regression |
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 |