CN-115599992-B - Product recommendation method, device and storage medium

CN115599992BCN 115599992 BCN115599992 BCN 115599992BCN-115599992-B

Abstract

The application relates to a product recommending method, a device and a storage medium, wherein the product recommending method comprises the steps of obtaining historical data of a product, constructing a first data matrix according to the historical data, carrying out similarity analysis on the first data matrix, combining function points with similarity meeting a set threshold according to a similarity analysis result into a first product function group, calculating information gain of the first product function group, selecting the first product function group according to the information gain to obtain a second product function group, calculating product recommending probability of the second product function group, and recommending products according to the product recommending probability. According to the application, the calculation amount of product data analysis in the post-step is greatly reduced, and the calculation efficiency is improved.

Inventors

WANG HUI
ZHENG JIANPENG
ZHOU XIAOMIN
KONG WEISHENG
DENG ZHIJI

Assignees

浙江大华技术股份有限公司

Dates

Publication Date: 20260508
Application Date: 20220926

Claims (6)

1. A method of product recommendation, the method comprising: Acquiring historical data of the product, and constructing a first data matrix according to the historical data, wherein the first data matrix comprises function points supported by the product; performing similarity analysis on the first data matrix, and combining the function points with the similarity meeting a set threshold value into a first product function group according to a similarity analysis result; calculating the information gain of the first product function group; Selecting the first product function group according to the information gain to obtain a second product function group; calculating the product recommendation probability of the second product function group, and recommending products according to the product recommendation probability; matrix elements of the first data matrix Representing whether a product i supports a function point j, wherein i is a natural number from 1 to M, j is a natural number from 1 to N, M is the number of products, and N is the number of the function points; performing similarity analysis on the first data matrix, and merging the function points with the similarity meeting a set threshold into a first product function group according to a similarity analysis result, wherein the method comprises the following steps: Calculating the similarity between column vectors of the first data matrix, merging column vectors with the similarity greater than a first preset value in the first data matrix, and generating a second data matrix, wherein the column vectors of the second data matrix comprise a plurality of first product function groups; Matrix elements of the second data matrix Indicating whether product i supports a first product function group j, where i is a natural number from 1 to M and j is from 1 to M The M is the product number, the natural number of For the number of first product function groups, each first product function group comprises one or more product function points; the historical data comprises characteristic data of the product; Calculating the information gain of the first product function group, comprising: generating a feature vector of the product according to the feature data of the product, wherein elements of the feature vector Features representing product i; Calculating a first entropy value according to the feature vector; calculating a second entropy value of the first product function group j according to the second data matrix; And calculating the information gain of each column vector of the second data matrix according to the first entropy value and the second entropy value.
2. The product recommendation method of claim 1, wherein selecting the first product function group according to the information gain, to obtain a second product function group comprises: sorting column vectors of the second data matrix according to the information gain; and selecting column vectors of which the information gains meet preset conditions in the second data matrix according to the sorting result, and generating the second product function group.
3. The product recommendation method of claim 2, wherein said calculating a product recommendation probability for said second product functionality comprises: and calculating the applicable probability and the inapplicable probability of the product according to the second product function group, and calculating the recommended probability of the product according to the applicable probability and the inapplicable probability.
4. A product recommendation method according to claim 3, wherein said calculating the probability of applicability and the probability of inapplicability of said product from said second set of product functions comprises: And calculating the applicability probability and inapplicability probability of the product by using a naive Bayesian algorithm according to the second product function group.
5. A product recommendation device, the device comprising: The system comprises an acquisition module, a first data matrix, a second data matrix and a display module, wherein the acquisition module is used for acquiring historical data of the product and constructing the first data matrix according to the historical data; The analysis module is used for carrying out similarity analysis on the first data matrix, and combining the function points with the similarity meeting a set threshold value into a first product function group according to a similarity analysis result; the calculation module is used for calculating the information gain of the first product function group; the selection module is used for selecting the first product function group according to the information gain to obtain a second product function group; the recommending module is used for calculating the product recommending probability of the second product functional group and recommending products according to the product recommending probability; matrix elements of the first data matrix Representing whether a product i supports a function point j, wherein i is a natural number from 1 to M, j is a natural number from 1 to N, M is the number of products, and N is the number of the function points; The analysis module is further configured to calculate a similarity between column vectors of the first data matrix, and combine column vectors with a similarity greater than a first preset value in the first data matrix to generate a second data matrix, where the column vectors of the second data matrix include a plurality of the first product function groups; Matrix elements of the second data matrix Indicating whether product i supports a first product function group j, where i is a natural number from 1 to M and j is from 1 to M The M is the product number, the natural number of For the number of first product function groups, each first product function group comprises one or more product function points; the historical data comprises characteristic data of the product; The computing module is also used for generating a characteristic vector of the product according to the characteristic data of the product, wherein the elements of the characteristic vector The method comprises the steps of representing the characteristics of a product i, calculating a first entropy value according to the characteristic vector, calculating a second entropy value of a first product function group j according to the second data matrix, and calculating information gain of each column vector of the second data matrix according to the first entropy value and the second entropy value.
6. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the product recommendation method according to any one of claims 1 to 4.

Description

Product recommendation method, device and storage medium Technical Field The present application relates to the field of data analysis technologies, and in particular, to a product recommendation method, device, and storage medium. Background The products in the security industry have the characteristics of multiple categories and abnormal functional characteristics. With the continuous division of product function research and development, products with similar overall but slight function differences are more and more, and the trouble is caused when customers use and purchase in practice. In addition, the security industry also has the characteristic of selecting products by industry clients, and the most suitable products of the same requirement are different aiming at different industry clients. Therefore, by analyzing the historical data of the product usage scenario of the customer, it is particularly important to recommend the most applicable product in the usage scenario for the customer. When recommending the most applicable product in the use scenario for the customer, a large amount of historical data of the use scenario of the customer's product needs to be analyzed. The data analysis mode used by the recommendation method in the prior art analyzes a large amount of historical data, so that the calculation and analysis efficiency is low. Aiming at the problem that the data analysis mode used in the prior art analyzes a large amount of historical data, so that the calculation and analysis efficiency is low, no effective solution is proposed at present. Disclosure of Invention The embodiment provides a product recommendation method, device and storage medium, which are used for solving the problem that the calculation and analysis efficiency is low because a large amount of historical data is analyzed in a data analysis mode used in the prior art. In a first aspect, in this embodiment, there is provided a product recommendation method, including: Acquiring historical data of the product, and constructing a first data matrix according to the historical data, wherein the first data matrix comprises function points supported by the product; performing similarity analysis on the first data matrix, and combining the function points with the similarity meeting a set threshold value into a first product function group according to a similarity analysis result; calculating the information gain of the first product function group; Selecting the first product function group according to the information gain to obtain a second product function group; and calculating the product recommendation probability of the second product function group, and recommending products according to the product recommendation probability. In some embodiments, the matrix element a ij of the first data matrix represents whether the product i supports the function point j, where i is a natural number from 1 to M, j is a natural number from 1 to N, M is the number of products, and N is the number of function points. In some embodiments, the performing similarity analysis on the first data matrix, and merging the function points with the similarity meeting the set threshold into the first product function group according to the similarity analysis result includes: And calculating the similarity between column vectors of the first data matrix, combining column vectors with the similarity larger than a first preset value in the first data matrix, and generating a second data matrix, wherein the column vectors of the second data matrix comprise a plurality of first product function groups. In some of these embodiments, said calculating the information gain of the first set of product functions comprises: and calculating the information gain of each column vector of the second data matrix. In some embodiments, the selecting the first product function group according to the information gain, and obtaining the second product function group includes: sorting column vectors of the second data matrix according to the information gain; and selecting column vectors of which the information gains meet preset conditions in the second data matrix according to the sorting result, and generating the second product function group. In some of these embodiments, the calculating the product recommendation probability for the second set of product functions includes: and calculating the applicable probability and the inapplicable probability of the product according to the second product function group, and calculating the recommended probability of the product according to the applicable probability and the inapplicable probability. In some of these embodiments, the calculating the probability of applicability and the probability of inapplicability of the product from the second set of product functions comprises: And calculating the applicability probability and inapplicability probability of the product by using a naive Bayesian algorithm according to