E-commerce comment voting social contact system based on neural network

A neural network and social system technology, applied in the field of product recommendation, can solve problems such as lack of rating information, selection of high-tech products, inability to help users without scientific and technological knowledge, etc., to ensure sales quality, achieve grasp, and improve product quality and service. Effect

Pending Publication Date: 2022-07-08
深圳市网睿科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to solve the problem that the existing system in the prior art cannot help users without scientific and technological knowledge choose high-tech products, and the recommendation system cannot effectively recommend new products due to the lack of scoring information. A social system for online e-commerce comment voting

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • E-commerce comment voting social contact system based on neural network
  • E-commerce comment voting social contact system based on neural network
  • E-commerce comment voting social contact system based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0055] refer to figure 1 , a social system for e-commerce comment voting based on neural network, including data collection system, data processing system and data application system, the output end of data collection system is connected with the input end of data processing system, the output end of data processing type is connected with data application The input terminal of the system is connected;

[0056] The data collection system includes the collection of raw data and the collection of user data;

[0057] The data processing system includes a self-organizing neural network model and a multi-feature word vector structure model, and the output end of the self-organizing neural network model and the input end of the multi-feature word vector structure model are linked through a contact matrix;

[0058] The data application system includes processing existing product data, predicting new product scores, capturing user shopping needs and recommending user shopping needs; ...

Embodiment 2

[0064] It has the implementation content of the above-mentioned embodiment, wherein, for the specific implementation of the above-mentioned embodiment, reference may be made to the above-mentioned description, and the embodiment here will not be repeated in detail; and in the embodiment of the present application, the difference between it and the above-mentioned embodiment is that :

[0065] The processing of the self-organizing neural network model includes the following steps:

[0066] Step 1: Use SOM model supervised learning to classify products according to the score value;

[0067] Step 2: Train the SOM model to score the existing product data;

[0068] Step 3: Then predict the new product data score.

[0069] In the present invention, the processing process of the multi-feature word vector structure model is to obtain effective local information and reduce the influence of noise data by implementing the attention mechanism to assign weights to the output codes of the...

Embodiment 3

[0071] It has the implementation content of the above-mentioned embodiment, wherein, for the specific implementation of the above-mentioned embodiment, reference may be made to the above-mentioned description, and the embodiment here will not be repeated in detail; and in the embodiment of the present application, the difference between it and the above-mentioned embodiment is that :

[0072] In step 1, the topology structure of the number of attributes of L products will be dynamically maintained,

[0073] ||C il -Wv BMU ||=min k {||C il -Wv k ||} (1.1)

[0074] Wv(t+1)=Wv(t)+θ(t)α(t)(C il (t)-Wv(t)) (1.2)

[0075] C in the above formula il Represents the target input vector of standard l of instance i, Wv represents the current weight vector, and BMU is the same as the input vector C il Neurons with nearest Euclidean distance, θ(t) is the domain function limited by the BMU distance, α(t) is the time constraint factor, and t is the current time cursor.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an e-commerce comment voting social contact system based on a neural network, which comprises a data acquisition system, a data processing system and a data application system, and is characterized in that the output end of the data acquisition system is connected with the input end of the data processing system, and the output end of the data processing system is connected with the input end of the data application system; the data acquisition system comprises acquisition of original data and acquisition of user data. According to the method, potential features of users are actively captured through a nonlinear network structure deeply learned by a social system, quantity data are mapped to a unified dimension, user behaviors and interests and preferences of the user behaviors are potentially associated automatically, and potential scores and possible recommendation conditions of users of products which are not online are predicted accordingly, so that the user experience is improved. Therefore, the product quality service can be adjusted in time, and the consumption psychology of the user is mastered.

Description

technical field [0001] The invention relates to the technical field of commodity recommendation, in particular to a neural network-based social system for e-commerce comment voting. Background technique [0002] With the continuous development of science and technology, the acceleration of the social informatization process, and the continuous improvement of e-commerce trading platforms, more and more people use online shopping to obtain the goods they need. The types of goods can be related to people's daily life. In all aspects, it provides great convenience for people's lives. Due to the wide variety of goods on various e-commerce platforms, users often spend a lot of time choosing to find the goods they need when shopping. Therefore, in order to ensure User experience, all provide personalized push services for users. [0003] There are two commonly used systems: passive recommendation and active recommendation. Active recommendation is based on the user's consumption,...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q30/06G06N3/04G06N3/08
CPCG06Q30/0631G06N3/08G06N3/045
Inventor 李宇飞李玉秀
Owner 深圳市网睿科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap