# Providing an Approach to Predicting Customer Quality in E-Commerce Social Networks Based on Big Data and Unsupervised Learning Method

arXiv: https://arxiv.org/abs/2109.02080 · HTML: https://mlsanity.com/paper/2109.02080/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 0 of its 47 references that are in the map.
- First posted on arXiv 5 Sep 2021.

## Abstract (arXiv)

One of the goals of every business enterprise is to increase customer loyalty. The degree of customer loyalty is called customer quality which its forecasting will affect strategic marketing practices. The purpose of this study is to predict the quality of customers of large e-commerce social networks by big data algorithms and unsupervised learning. For this purpose, a graph-based social network analysis framework was used for community detection in the Stanford Network Analysis Platform (SNAP). Then in the found communities, the quality of customers was predicted. The results showed that various visits with an impact of 37.13% can have the greatest impact on customer quality and the order of impact of other parameters were from highest to lowest: number of frequent customer visits (28.56%), role in social networks (28.37%), Indirect transactions (26.74%), activity days (25.62%) and customer social network size (25.06%).