CN116703520A

CN116703520A - Product recommendation method based on improved K-means algorithm and related equipment thereof

Info

Publication number: CN116703520A
Application number: CN202310713618.1A
Authority: CN
Inventors: 刘兴廷; 王鹏梅
Original assignee: Ping An Property and Casualty Insurance Company of China Ltd
Current assignee: Ping An Property and Casualty Insurance Company of China Ltd
Priority date: 2023-06-15
Filing date: 2023-06-15
Publication date: 2023-09-05

Abstract

The embodiment of the application belongs to the technical field of financial science and technology, is applied to the field of intelligent product recommendation in a group purchase scene, and relates to a product recommendation method based on an improved K-means algorithm and related equipment thereof, wherein the method comprises the steps of obtaining evaluation results of all users in a target user set on different products respectively; clustering users with the same evaluation result level of the same product together through twice K-means clustering treatment, and according to the user feature matrixes of all users in the same clustering center; constructing a target feature matrix corresponding to a user set to be recommended according to feature values of all the users to be recommended in the user set to be recommended; and screening target products from the different products according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm, and recommending the target products to the user set to be recommended. The method and the system realize the efficient recommendation of the target product with high evaluation value from a plurality of products for a new group purchase group in an electronic shopping scene.

Description

Product recommendation method based on improved K-means algorithm and related equipment thereof

Technical Field

The application relates to the technical field of finance and technology, is applied to the field of intelligent product recommendation in a group purchase scene, and particularly relates to a product recommendation method based on an improved K-means algorithm and related equipment thereof.

Background

With the rapid development of internet technology and electronic information technology, the information generation and propagation speed is greatly accelerated, and meanwhile, huge data makes it difficult for people to screen out useful data information. In order to help users to obtain needed information from massive information with high efficiency, scholars at home and abroad propose a few schemes, wherein the most common is a method for using a search engine. Search engines require users to provide accurate desired keywords and it is difficult to obtain useful information if the keywords are inaccurate. To solve this problem, recommendation systems have been developed. The existing recommendation methods mostly use a deep learning algorithm to train a recommendation model for product recommendation, and the methods need to establish a complex data model and huge sample data, so that the difficulty is high.

Aiming at a product recommendation scene when a user performs electronic shopping, the complexity brought by huge data samples to classification needs to be considered. In addition, most of the existing group purchase mode product recommendation modes are that all users firstly act on products to select and obtain a desired product, and then accurately select the product; however, this method has a certain limitation, needs to be contacted with the group purchase user in advance, and is not suitable for the situation that the group purchase group is not specific or the e-commerce platform actively recommends the group purchase product. Therefore, when the prior art performs group purchase, under the conditions that the quality of multiple products is not known and the purchase intention of group purchase personnel on target products cannot be clarified, the problem that high-quality products cannot be efficiently recommended to the group purchase group exists.

Disclosure of Invention

The embodiment of the application aims to provide a product recommendation method based on an improved K-means algorithm and related equipment thereof, which are used for solving the problem that high-quality products cannot be efficiently recommended to a group purchase group under the conditions that the quality of multiple products is unknown and the purchase intention of group purchase personnel on target products cannot be clearly determined in the prior art.

In order to solve the technical problems, the embodiment of the application provides a product recommendation method based on an improved K-means algorithm, which adopts the following technical scheme:

a product recommendation method based on an improved K-means algorithm comprises the following steps:

obtaining evaluation results of all users in the target user set on different products respectively;

clustering the evaluation results of the different products by all users according to an improved K-means algorithm to obtain a final clustering result;

acquiring user characteristic information of each target user belonging to the same cluster center according to the final cluster processing result, and constructing a user characteristic matrix corresponding to the same cluster center according to the user characteristic information of each target user of the same cluster center;

Repeatedly executing the user characteristic matrix construction step to obtain user characteristic matrixes corresponding to different clustering centers;

acquiring and calculating the characteristic values of all users to be recommended in a preset user set to be recommended according to the user characteristic information of the users to be recommended, and constructing a target characteristic matrix corresponding to the user set to be recommended according to the characteristic values of all the users to be recommended;

and screening target products from the different products according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm, and recommending the target products to the user set to be recommended.

Further, the step of obtaining the evaluation results of all the users in the target user set on different products respectively specifically includes:

acquiring user distinguishing identification information of all users in the target user set, wherein the user distinguishing identification information comprises user IDs;

acquiring product distinguishing identification information of different products, wherein the product distinguishing identification information comprises product distinguishing names;

inquiring and extracting evaluation results of all users in the target user set on different products according to the user distinguishing identification information and the product distinguishing identification information;

And caching the user distinguishing identification information, the product distinguishing identification information and the extracted corresponding evaluation result in a triplet form.

Further, before executing the step of clustering the evaluation results of the different products by all the users according to the improved K-means algorithm to obtain a final clustering result, the method further includes:

acquiring cached data of all triples, and counting the number of all triples;

constructing data points with the same number as all triples in a preset clustering space;

assigning data in different triples in all triples to different data points one by one, and constructing a one-to-one correspondence for the data points and the triples;

the step of clustering the evaluation results of the different products by all users according to the improved K-means algorithm to obtain a final clustering result comprises the following steps:

according to the product distinguishing identification information in the triples and the improved K-means algorithm, performing primary clustering treatment on the evaluation results of different products by all users respectively to obtain an initial clustering result;

Performing secondary clustering on the initial clustering result according to different evaluation results of all users on the same product and the improved K-means algorithm to obtain a secondary clustering result;

and taking the secondary clustering result as the final clustering result.

Further, the step of performing primary clustering processing on the evaluation results of the different products by all the users according to the product distinguishing identification information in the triplet and the improved K-means algorithm to obtain an initial clustering result specifically includes:

counting the quantity of different product distinguishing identification information in the triples;

setting the number of different product distinguishing identification information in the triplet as the number of clustering centers of the improved K-means algorithm;

according to the different product distinguishing identification information, carrying out clustering arrangement on the triples, and arranging all triples of the same product distinguishing identification information into an initial aggregation set;

and respectively acquiring initial clustering sets corresponding to different product distinguishing identification information as initial clustering results to finish initial clustering processing.

Further, before the step of performing the secondary clustering process on the initial clustering result according to the different evaluation results of all users on the same product and the improved K-means algorithm to obtain a secondary clustering result, the method further includes:

Respectively carrying out numerical processing on the evaluation results in different initial clustering sets to obtain numerical evaluation values respectively corresponding to the elements in different initial clustering sets;

the step of performing secondary clustering treatment on the initial clustering result according to different evaluation results of the same product by all users and the improved K-means algorithm to obtain a secondary clustering treatment result specifically comprises the following steps:

acquiring a preset evaluation value division layer number according to a preset hierarchy division rule;

setting the preset evaluation value division layer number as the number of clustering centers of the improved K-means algorithm;

according to the numerical evaluation values respectively corresponding to the elements in the same initial cluster set, carrying out cluster arrangement on the triples in the same initial cluster set, and arranging the triples in the same evaluation value division level into a secondary cluster set;

and respectively acquiring secondary clustering sets in different initial clustering sets as secondary clustering processing results to finish the secondary clustering processing.

Further, the step of obtaining the user characteristic information of each target user belonging to the same cluster center according to the final cluster processing result specifically includes:

Optionally a secondary cluster set as the same cluster center;

analyzing the triple in the secondary cluster set to obtain all analyzed user distinguishing identification information;

acquiring user characteristic information of a target user from a preset user characteristic information table according to the analyzed all user distinguishing identification information, wherein the user characteristic information comprises user age, user ID, product preference information and user gender;

the step of constructing the user feature matrix corresponding to the same cluster center according to the user feature information of each target user in the same cluster center specifically comprises the following steps:

acquiring weight values preset for different user characteristic information respectively;

according to the weight values preset for different user characteristic information and the user characteristic information corresponding to the current target user, accumulating and summing operation is carried out, and the characteristic value of the current target user is calculated;

and acquiring the characteristic values of all target users in the same cluster center, performing determinant display on the characteristic values of all target users in the same cluster center according to a preset matrix, and constructing and completing the user characteristic matrix.

Further, the step of obtaining and calculating the feature values of all the users to be recommended in the preset set of users to be recommended according to the user feature information of the users to be recommended, and constructing a target feature matrix corresponding to the set of users to be recommended according to the feature values of all the users to be recommended specifically includes:

acquiring weight values respectively corresponding to the user characteristic information of the current user to be recommended;

performing accumulation operation on the weight values corresponding to the user characteristic information respectively, and calculating the characteristic value of the current user to be recommended;

acquiring the characteristic value of each user to be recommended in the user set to be recommended, carrying out determinant display on the characteristic value of each user to be recommended in the user set to be recommended according to a preset matrix, and constructing and completing the target characteristic matrix;

the step of screening target products from the different products to recommend the target products to the user set to be recommended according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm specifically comprises the following steps:

according to a preset similarity algorithm, calculating the similarity between the target feature matrix and the user feature matrix corresponding to the different clustering centers respectively;

Screening the first K user feature matrixes with highest similarity from the user feature matrixes corresponding to different clustering centers according to a preset K neighbor algorithm and the similarity, wherein K is a positive integer;

determining secondary aggregation sets respectively corresponding to the first K user feature matrixes with the highest similarity;

screening secondary clustering sets when the evaluation value is the highest value from the secondary clustering sets respectively corresponding to the first K user feature matrixes with the highest similarity;

and determining an initial aggregation set to which the secondary aggregation set belongs when the evaluation value is the highest value, acquiring a product corresponding to the initial aggregation set as a target product, and pushing the target product to each user in the user set to be recommended.

In order to solve the technical problems, the embodiment of the application also provides a product recommendation device based on an improved K-means algorithm, which adopts the following technical scheme:

a product recommendation device based on an improved K-means algorithm, comprising:

the evaluation result acquisition module is used for acquiring the evaluation results of all users in the target user set on different products respectively;

the K-means clustering module is used for carrying out clustering processing on the evaluation results of the different products by all users according to an improved K-means algorithm to obtain a final clustering processing result;

The user characteristic matrix construction module is used for acquiring the user characteristic information of each target user belonging to the same cluster center according to the final cluster processing result, and constructing a user characteristic matrix corresponding to the same cluster center according to the user characteristic information of each target user of the same cluster center;

the repeated execution module is used for repeatedly executing the user characteristic matrix construction step and obtaining user characteristic matrixes corresponding to different clustering centers;

the target feature matrix construction module is used for acquiring and calculating the feature values of all the users to be recommended in a preset user set to be recommended according to the user feature information of the users to be recommended, and constructing a target feature matrix corresponding to the user set to be recommended according to the feature values of all the users to be recommended;

and the target product screening module is used for screening target products from different products according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm, and recommending the target products to the user set to be recommended.

In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical schemes:

A computer device comprising a memory having stored therein computer readable instructions which when executed by the processor implement the steps of the product recommendation method described above based on the modified K-means algorithm.

In order to solve the above technical problems, an embodiment of the present application further provides a computer readable storage medium, which adopts the following technical schemes:

a computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of a product recommendation method based on the modified K-means algorithm as described above.

Compared with the prior art, the embodiment of the application has the following main beneficial effects:

according to the product recommendation method based on the improved K-means algorithm, the evaluation results of all users in a target user set on different products are obtained; clustering the evaluation results of the different products by all users according to an improved K-means algorithm to obtain a final clustering result; acquiring user characteristic information of each target user belonging to the same cluster center according to the final cluster processing result, and constructing a user characteristic matrix corresponding to the same cluster center according to the user characteristic information of each target user of the same cluster center; repeatedly executing the user characteristic matrix construction step to obtain user characteristic matrixes corresponding to different clustering centers; acquiring and calculating the characteristic values of all users to be recommended in a preset user set to be recommended according to the user characteristic information of the users to be recommended, and constructing a target characteristic matrix corresponding to the user set to be recommended according to the characteristic values of all the users to be recommended; and screening target products from the different products according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm, and recommending the target products to the user set to be recommended. And clustering users with the same evaluation result level of the same product together through twice K-means clustering treatment, and recommending target products with high evaluation values from a plurality of products for a new group purchase group efficiently according to user feature matrixes of all users in the same clustering center.

Drawings

In order to more clearly illustrate the solution of the present application, a brief description will be given below of the drawings required for the description of the embodiments of the present application, it being apparent that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained from these drawings without the exercise of inventive effort for a person of ordinary skill in the art.

FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;

FIG. 2 is a flow chart of one embodiment of a product recommendation method based on the modified K-means algorithm in accordance with the present application;

FIG. 3 is a flow chart of one embodiment of step 201 of FIG. 2;

FIG. 4 is a flow chart of one embodiment of step 202 of FIG. 2;

FIG. 5 is a flow chart of one embodiment of step 401 shown in FIG. 4;

FIG. 6 is a flow chart of one embodiment of step 402 shown in FIG. 4;

FIG. 7 is a flow chart of one embodiment of step 206 of FIG. 2;

FIG. 8 is a schematic diagram of one embodiment of a product recommendation device based on the modified K-means algorithm in accordance with the present application;

FIG. 9 is a schematic diagram of a configuration of one embodiment of 802 shown in FIG. 8;

FIG. 10 is a schematic diagram of an embodiment of a computer device in accordance with the present application.

Detailed Description

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the applications herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description of the application and the claims and the description of the drawings above are intended to cover a non-exclusive inclusion. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

In order to make the person skilled in the art better understand the solution of the present application, the technical solution of the embodiment of the present application will be clearly and completely described below with reference to the accompanying drawings.

As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or send messages or the like. Various communication client applications, such as a web browser application, a shopping class application, a search class application, an instant messaging tool, a mailbox client, social platform software, etc., may be installed on the terminal devices 101, 102, 103.

The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, electronic book readers, MP3 players (Moving Picture ExpertsGroup Audio Layer III, dynamic video expert compression standard audio plane 3), MP4 (Moving PictureExperts Group Audio Layer IV, dynamic video expert compression standard audio plane 4) players, laptop and desktop computers, and the like.

The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the terminal devices 101, 102, 103.

It should be noted that, the product recommendation method based on the improved K-means algorithm provided by the embodiment of the present application is generally executed by a server/terminal device, and correspondingly, the product recommendation device based on the improved K-means algorithm is generally disposed in the server/terminal device.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

With continued reference to FIG. 2, a flowchart of one embodiment of a product recommendation method based on the modified K-means algorithm in accordance with the present application is shown. The product recommendation method based on the improved K-means algorithm comprises the following steps:

step 201, obtaining evaluation results of all users in the target user set on different products respectively.

With continued reference to fig. 3, fig. 3 is a flow chart of one embodiment of step 201 of fig. 2, comprising:

step 301, obtaining user distinguishing identification information of all users in the target user set, wherein the user distinguishing identification information comprises user IDs;

Step 302, obtaining product distinguishing identification information of different products, wherein the product distinguishing identification information comprises product distinguishing names;

step 303, inquiring and extracting the evaluation results of all users in the target user set on different products according to the user distinguishing identification information and the product distinguishing identification information;

and step 304, caching the user distinguishing identification information, the product distinguishing identification information and the extracted corresponding evaluation result in a triplet form.

By caching the user distinguishing identification information, the product distinguishing identification information and the extracted corresponding evaluation results in the form of triples, the evaluation results of different users on different products can be more intuitively represented, and the corresponding data content can be conveniently reused later by caching in the form of triples.

And 202, clustering the evaluation results of the different products by all users according to an improved K-means algorithm to obtain a final clustering result.

In this embodiment, before executing the step of clustering the evaluation results of the different products by the all users according to the modified K-means algorithm to obtain a final clustering result, the method further includes: acquiring cached data of all triples, and counting the number of all triples; constructing data points with the same number as all triples in a preset clustering space; and assigning data in different triples in all triples to different data points one by one, and constructing a one-to-one correspondence relationship for the data points and the triples.

By constructing data points with the same number as all triples in a preset clustering space and assigning the data points in different triples in all triples to different data points one by one, the method realizes that all triples are converted into discrete point value information distributed in a certain space, and is convenient for improving the usability of a K-means clustering algorithm.

With continued reference to FIG. 4, FIG. 4 is a flow chart of one embodiment of step 202 of FIG. 2, including:

step 401, performing primary clustering treatment on the evaluation results of different products by all users according to the product distinguishing identification information in the triplet and the improved K-means algorithm, and obtaining an initial clustering result;

with continued reference to fig. 5, fig. 5 is a flow chart of one embodiment of step 401 shown in fig. 4, comprising:

step 501, counting the number of different product distinguishing identification information in the triples;

step 502, setting the number of different product distinguishing identification information in the triplet as the number of clustering centers of the improved K-means algorithm;

step 503, according to the different product distinguishing identification information, performing clustering arrangement on the triples, and arranging all triples of the same product distinguishing identification information into an initial aggregation set;

And step 504, respectively acquiring initial clustering sets corresponding to different product distinguishing identification information as initial clustering results to finish initial clustering processing.

Firstly, clustering the evaluation results corresponding to different products for the first time by taking the number of the products as a clustering center, and clustering the evaluation results corresponding to the same products together. And the secondary clustering and analysis of the subsequent program are facilitated.

Step 402, performing secondary clustering on the initial clustering result according to different evaluation results of the same product by all users and the improved K-means algorithm to obtain a secondary clustering result;

in this embodiment, before the step of performing the secondary clustering process on the initial clustering result according to the different evaluation results of all users on the same product and the improved K-means algorithm to obtain a secondary clustering result, the method further includes: and respectively carrying out numerical processing on the evaluation results in different initial clustering sets to obtain numerical evaluation values respectively corresponding to the elements in different initial clustering sets.

With continued reference to fig. 6, fig. 6 is a flow chart of one embodiment of step 402 shown in fig. 4, comprising:

step 601, obtaining a preset evaluation value division layer number according to a preset hierarchy division rule;

In this embodiment, the preset hierarchical division rule may be a set high, medium, and low evaluation level, or may be an evaluation value interval, for example: the numerical range of the evaluation value is [0,1], and the numerical range is divided into 5 hierarchical ranges of [0,0.2 ], [0.2,0.4 ], [0.4,0.6 ], [0.6,0.8) and [0.8,1], respectively.

Step 602, setting the preset evaluation value division layer number as the number of clustering centers of the improved K-means algorithm;

step 603, according to the numerical evaluation values respectively corresponding to the elements in the same initial cluster set, carrying out cluster arrangement on the triples in the same initial cluster set, and arranging the triples in the same evaluation value division level into a secondary cluster set;

step 604, respectively obtaining secondary clustering sets in different initial clustering sets as secondary clustering processing results to finish the secondary clustering processing.

And in the second clustering, dividing the evaluation results obtained in the first clustering according to the evaluation levels, and carrying out clustering treatment, namely, respectively clustering the evaluation results of different evaluation levels together. And the machine learning and analysis of the subsequent program according to the clustering result are facilitated.

And step 403, taking the secondary clustering result as the final clustering result.

Through twice K-means clustering treatment, firstly, clustering the evaluation results corresponding to different products for the first time by taking the number of the products as a clustering center, and clustering the evaluation results corresponding to the same products together; and then, in the second clustering, dividing the evaluation results obtained in the first clustering according to the evaluation level, and carrying out clustering treatment, namely, respectively clustering the evaluation results of high evaluation values and low evaluation values together. Through the clustering process for two times, the user groups with high evaluation values and low evaluation values are distinguished when each product is evaluated.

Step 203, obtaining user feature information of each target user belonging to the same cluster center according to the final cluster processing result, and constructing a user feature matrix corresponding to the same cluster center according to the user feature information of each target user of the same cluster center.

In this embodiment, the step of obtaining the user feature information of each target user belonging to the same cluster center according to the final cluster processing result specifically includes: optionally a secondary cluster set as the same cluster center; analyzing the triple in the secondary cluster set to obtain all analyzed user distinguishing identification information; and acquiring user characteristic information of the target user from a preset user characteristic information table according to the analyzed all user distinguishing identification information, wherein the user characteristic information comprises user age, user ID, product preference information and user gender.

In this embodiment, the step of constructing the user feature matrix corresponding to the same cluster center according to the user feature information of each target user in the same cluster center specifically includes: acquiring weight values preset for different user characteristic information respectively; according to the weight values preset for different user characteristic information and the user characteristic information corresponding to the current target user, accumulating and summing operation is carried out, and the characteristic value of the current target user is calculated; and acquiring the characteristic values of all target users in the same cluster center, performing determinant display on the characteristic values of all target users in the same cluster center according to a preset matrix, and constructing and completing the user characteristic matrix.

And acquiring user characteristic information of each user in the same clustering center through a clustering result and corresponding triple data, constructing a characteristic value corresponding to each user according to a weight value preset for the characteristic information of each user, and performing determinant display on the characteristic values of all target users in the same clustering center by using a matrix to construct and finish the user characteristic matrix. And constructing a feature matrix of the users clustered in the same evaluation result level, so that the subsequent program can realize high-value product recommendation by directly using the user feature information of the latest user.

And 204, repeatedly executing the user feature matrix construction step to obtain the user feature matrices corresponding to different clustering centers.

Through step 203 and step 204, user feature matrixes corresponding to different clustering centers are obtained, feature matrix construction is carried out on users clustered in the same evaluation result level, and high-value product recommendation can be realized by a subsequent program directly through the user feature information of the latest user.

Step 205, obtaining and calculating the characteristic values of all the users to be recommended in a preset user set to be recommended according to the user characteristic information of the users to be recommended, and constructing a target characteristic matrix corresponding to the user set to be recommended according to the characteristic values of all the users to be recommended.

In this embodiment, the step of obtaining and calculating feature values of all users to be recommended in a preset set of users to be recommended according to user feature information of the users to be recommended, and constructing a target feature matrix corresponding to the set of users to be recommended according to the feature values of all the users to be recommended specifically includes: acquiring weight values respectively corresponding to the user characteristic information of the current user to be recommended; performing accumulation operation on the weight values corresponding to the user characteristic information respectively, and calculating the characteristic value of the current user to be recommended; acquiring the characteristic value of each user to be recommended in the user set to be recommended, carrying out determinant display on the characteristic value of each user to be recommended in the user set to be recommended according to a preset matrix, and constructing and completing the target characteristic matrix.

And 206, screening target products from the different products according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm, and recommending the target products to the user set to be recommended.

With continued reference to fig. 7, fig. 7 is a flow chart of one embodiment of step 206 shown in fig. 2, comprising:

step 701, calculating the similarity between the target feature matrix and the user feature matrix corresponding to the different clustering centers respectively according to a preset similarity algorithm;

step 702, screening the first K user feature matrices with highest similarity from the user feature matrices corresponding to the different clustering centers according to a preset K neighbor algorithm and the similarity, wherein K is a positive integer;

step 703, determining secondary aggregation sets corresponding to the first K user feature matrices with the highest similarity respectively;

step 704, screening out secondary clustering sets with the highest evaluation value from the secondary clustering sets respectively corresponding to the first K user feature matrixes with the highest similarity;

step 705, determining an initial cluster set to which the secondary cluster set belongs when the evaluation value is the highest value, obtaining a product corresponding to the initial cluster set as a target product, and pushing the target product to each user in the user set to be recommended.

By adopting the improved K-means clustering algorithm and the K nearest neighbor algorithm, the problems that when the group purchase is carried out, the purchase will of group purchase personnel on target products is uneven and the quality of the final purchased products cannot be guaranteed under the condition that the quality of multiple products is not known are solved. The method and the device can be used for efficiently screening high-quality target products from multiple products for multiple group purchase users in a group purchase scene.

In the foregoing step 205 and step 206, when the actual application scenario is that the company sends benefit for staff, benefit item collection is performed, and at this time, benefit product screening is performed by using feature matrices corresponding to all staff in the company, which represents efficient screening of target products for multiple users from multiple products.

In addition, there are very special cases, for example, a case that only one user to be recommended exists in the preset set of users to be recommended, at this time, step 205 is executed to obtain a feature value of the user to be recommended, and then the feature value is used as a search field to identify the frequency of occurrence of the feature value in the user feature matrices corresponding to the different clustering centers; screening out K user feature matrixes before ranking according to the frequency; determining secondary aggregation sets respectively corresponding to the first K user feature matrixes; screening secondary clustering sets when the evaluation value is the highest value from the secondary clustering sets respectively corresponding to the first K user feature matrixes; and determining an initial aggregation set to which the secondary aggregation set belongs when the evaluation value is the highest value, acquiring a product corresponding to the initial aggregation set as a target product, and pushing the target product to the user to be recommended.

According to the method, the evaluation results of all users in the target user set on different products are obtained; clustering the evaluation results of the different products by all users according to an improved K-means algorithm to obtain a final clustering result; acquiring user characteristic information of each target user belonging to the same cluster center according to the final cluster processing result, and constructing a user characteristic matrix corresponding to the same cluster center according to the user characteristic information of each target user of the same cluster center; repeatedly executing the user characteristic matrix construction step to obtain user characteristic matrixes corresponding to different clustering centers; acquiring and calculating the characteristic values of all users to be recommended in a preset user set to be recommended according to the user characteristic information of the users to be recommended, and constructing a target characteristic matrix corresponding to the user set to be recommended according to the characteristic values of all the users to be recommended; and screening target products from the different products according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm, and recommending the target products to the user set to be recommended. And clustering users with the same evaluation result level of the same product together through twice K-means clustering treatment, and recommending target products with high evaluation values from a plurality of products for a new group purchase group efficiently according to user feature matrixes of all users in the same clustering center.

The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

In the embodiment of the application, users with the same evaluation result level of the same product are clustered together through twice K-means clustering processing, and target products with high evaluation values are recommended from a plurality of products for a new group purchase group with high efficiency according to the user characteristic matrixes of all users in the same clustering center.

With further reference to fig. 8, as an implementation of the method shown in fig. 2, the present application provides an embodiment of a product recommendation device based on the modified K-means algorithm, where the embodiment of the device corresponds to the embodiment of the method shown in fig. 2, and the device is specifically applicable to various electronic devices.

As shown in fig. 8, the product recommendation device 800 based on the modified K-means algorithm according to the present embodiment includes: the device comprises an evaluation result acquisition module 801, a K-means clustering module 802, a user feature matrix construction module 803, a repeated execution module 804, a target feature matrix construction module 805 and a target product screening module 806. Wherein:

an evaluation result obtaining module 801, configured to obtain evaluation results of all users in the target user set on different products respectively;

the K-means clustering module 802 is configured to perform clustering on the evaluation results of the different products for all the users according to an improved K-means algorithm, so as to obtain a final clustering result;

the user feature matrix construction module 803 is configured to obtain user feature information of each target user belonging to the same cluster center according to the final cluster processing result, and construct a user feature matrix corresponding to the same cluster center according to the user feature information of each target user of the same cluster center;

the repeated execution module 804 is configured to repeatedly execute the user feature matrix construction step to obtain user feature matrices corresponding to different clustering centers;

the target feature matrix construction module 805 is configured to obtain and calculate feature values of all users to be recommended in a preset set of users to be recommended according to user feature information of the users to be recommended, and construct a target feature matrix corresponding to the set of users to be recommended according to the feature values of all the users to be recommended;

And a target product screening module 806, configured to screen target products from the different products according to a target feature matrix corresponding to the user set to be recommended, user feature matrices corresponding to the different clustering centers, and a preset K nearest neighbor algorithm, so as to recommend the target products to the user set to be recommended.

With continued reference to fig. 9, fig. 9 is a schematic diagram of a structure of a specific embodiment of the module 802 shown in fig. 8, where the K-means clustering module 802 includes a first clustering sub-module 802a and a second clustering sub-module 802b. The first clustering submodule 802a includes a first statistics unit 901, a first setting unit 902 for the number of cluster centers, an initial cluster collection sorting unit 903 and an initial cluster result obtaining unit 904, and the second clustering submodule 802b includes an evaluation level obtaining unit 905, a second setting unit 906 for the number of cluster centers, a secondary cluster collection sorting unit 907 and a secondary cluster result obtaining unit 908, where:

a first statistics unit 901, configured to count the number of different product distinguishing identifier information in the triples;

a first setting unit 902 of cluster center number, configured to set the number of different product distinguishing identification information in the triplet as the number of cluster centers of the modified K-means algorithm;

The initial cluster set arrangement unit 903 is configured to perform cluster arrangement on the triples according to the different product identification information, and arrange all triples of the same product identification information into an initial cluster set;

an initial clustering result obtaining unit 904, configured to obtain initial clustering sets corresponding to different product distinguishing identifier information, as initial clustering results, and complete initial clustering processing;

an evaluation level obtaining unit 905, configured to obtain a preset evaluation value division number of layers according to a preset level division rule;

a second setting unit 906 for setting the preset number of evaluation value division layers as the number of cluster centers of the improved K-means algorithm;

a secondary cluster set arrangement unit 907, configured to perform cluster arrangement on each triplet in the same initial cluster set according to the numeric evaluation values corresponding to each element in the same initial cluster set, and arrange the triples in the same evaluation value division level into a secondary cluster set;

and a secondary clustering result obtaining unit 908, configured to obtain secondary clustering sets in different initial clustering sets, as secondary clustering results, and complete the secondary clustering process.

Those skilled in the art will appreciate that implementing all or part of the above described embodiment methods may be accomplished by computer readable instructions, stored on a computer readable storage medium, that the program when executed may comprise the steps of embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 10, fig. 10 is a basic structural block diagram of a computer device according to the present embodiment.

The computer device 10 includes a memory 10a, a processor 10b, and a network interface 10c communicatively coupled to each other via a system bus. It should be noted that only computer device 10 having components 10a-10c is shown in the figures, but it should be understood that not all of the illustrated components need be implemented and that more or fewer components may alternatively be implemented. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculations and/or information processing in accordance with predetermined or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable gate arrays (fields-Programmable Gate Array, FPGAs), digital processors (Digital Signal Processor, DSPs), embedded devices, etc.

The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.

The memory 10a includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage 10a may be an internal storage unit of the computer device 10, such as a hard disk or a memory of the computer device 10. In other embodiments, the memory 10a may also be an external storage device of the computer device 10, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 10. Of course, the memory 10a may also include both internal storage units of the computer device 10 and external storage devices thereof. In this embodiment, the memory 10a is generally used to store an operating system and various types of application software installed on the computer device 10, such as computer readable instructions for a product recommendation method based on a modified K-means algorithm. Further, the memory 10a may be used to temporarily store various types of data that have been output or are to be output.

The processor 10b may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 10b is generally used to control the overall operation of the computer device 10. In this embodiment, the processor 10b is configured to execute computer readable instructions stored in the memory 10a or process data, such as computer readable instructions for executing the product recommendation method based on the modified K-means algorithm.

The network interface 10c may comprise a wireless network interface or a wired network interface, the network interface 10c typically being used to establish a communication connection between the computer device 10 and other electronic devices.

The computer equipment provided by the embodiment belongs to the technical field of financial science and technology, and is applied to the field of intelligent product recommendation in a group purchase scene. According to the method, the evaluation results of all users in the target user set on different products are obtained; clustering the evaluation results of the different products by all users according to an improved K-means algorithm to obtain a final clustering result; acquiring user characteristic information of each target user belonging to the same cluster center according to the final cluster processing result, and constructing a user characteristic matrix corresponding to the same cluster center according to the user characteristic information of each target user of the same cluster center; repeatedly executing the user characteristic matrix construction step to obtain user characteristic matrixes corresponding to different clustering centers; acquiring and calculating the characteristic values of all users to be recommended in a preset user set to be recommended according to the user characteristic information of the users to be recommended, and constructing a target characteristic matrix corresponding to the user set to be recommended according to the characteristic values of all the users to be recommended; and screening target products from the different products according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm, and recommending the target products to the user set to be recommended. And clustering users with the same evaluation result level of the same product together through twice K-means clustering treatment, and recommending target products with high evaluation values from a plurality of products for a new group purchase group efficiently according to user feature matrixes of all users in the same clustering center.

The present application also provides another embodiment, namely, a computer readable storage medium storing computer readable instructions executable by a processor to cause the processor to perform the steps of the product recommendation method based on the modified K-means algorithm as described above.

The computer readable storage medium provided by the embodiment belongs to the technical field of financial science and technology, and is applied to the field of intelligent product recommendation in a group purchase scene. According to the method, the evaluation results of all users in the target user set on different products are obtained; clustering the evaluation results of the different products by all users according to an improved K-means algorithm to obtain a final clustering result; acquiring user characteristic information of each target user belonging to the same cluster center according to the final cluster processing result, and constructing a user characteristic matrix corresponding to the same cluster center according to the user characteristic information of each target user of the same cluster center; repeatedly executing the user characteristic matrix construction step to obtain user characteristic matrixes corresponding to different clustering centers; acquiring and calculating the characteristic values of all users to be recommended in a preset user set to be recommended according to the user characteristic information of the users to be recommended, and constructing a target characteristic matrix corresponding to the user set to be recommended according to the characteristic values of all the users to be recommended; and screening target products from the different products according to the target feature matrix corresponding to the user set to be recommended, the user feature matrix corresponding to the different clustering centers and a preset K neighbor algorithm, and recommending the target products to the user set to be recommended. And clustering users with the same evaluation result level of the same product together through twice K-means clustering treatment, and recommending target products with high evaluation values from a plurality of products for a new group purchase group efficiently according to user feature matrixes of all users in the same clustering center.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.

It is apparent that the above-described embodiments are only some embodiments of the present application, but not all embodiments, and the preferred embodiments of the present application are shown in the drawings, which do not limit the scope of the patent claims. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a thorough and complete understanding of the present disclosure. Although the application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing description, or equivalents may be substituted for elements thereof. All equivalent structures made by the content of the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the scope of the application.

Claims

1. The product recommendation method based on the improved K-means algorithm is characterized by comprising the following steps of:

2. The product recommendation method based on the improved K-means algorithm according to claim 1, wherein the step of obtaining the evaluation results of all users in the target user set on different products respectively comprises the following steps:

3. The improved K-means algorithm based product recommendation method according to claim 2, wherein before the step of performing the clustering process on the evaluation results of the different products by the all users according to the improved K-means algorithm, respectively, to obtain a final clustering result, the method further comprises:

Acquiring cached data of all triples, and counting the number of all triples;

and taking the secondary clustering result as the final clustering result.

4. The method for recommending products based on the modified K-means algorithm according to claim 3, wherein the step of performing primary clustering on the evaluation results of the different products by all the users according to the product distinguishing identification information in the triplet and the modified K-means algorithm to obtain an initial clustering result specifically comprises:

5. The improved K-means algorithm based product recommendation method according to claim 3 or 4, wherein, before performing the step of performing the secondary clustering on the initial clustering result according to different evaluation results of the same product by all users and the improved K-means algorithm to obtain a secondary clustering result, the method further comprises:

6. The product recommendation method based on the improved K-means algorithm according to claim 5, wherein the step of obtaining user feature information of each target user belonging to the same cluster center according to the final cluster processing result specifically comprises:

optionally a secondary cluster set as the same cluster center;

7. The method for recommending products based on the improved K-means algorithm according to claim 6, wherein the step of obtaining and calculating the feature values of all users to be recommended in a preset set of users to be recommended according to the user feature information of the users to be recommended, and constructing the target feature matrix corresponding to the set of users to be recommended according to the feature values of all the users to be recommended specifically comprises the steps of:

8. A product recommendation device based on an improved K-means algorithm, comprising:

9. A computer device comprising a memory having stored therein computer readable instructions which when executed implement the steps of the product recommendation method based on the modified K-means algorithm of any of claims 1 to 7.

10. A computer readable storage medium, characterized in that it has stored thereon computer readable instructions which, when executed by a processor, implement the steps of the product recommendation method based on the modified K-means algorithm according to any of claims 1 to 7.