计算机与数字工程Issue(9)：2176-2179,2226,5.DOI:10.3969/j.issn.1672-9722.2019.09.014
基于深度学习BCCM模型的网上用户画像识别分析∗
Analysis of Online User Portrait Recognition Based on Deep Learning BCCM Model
摘要
Abstract
A BCCM model for identifying perceptual investors by analyzing content and behavioral characteristics is construct?ed by integrating traditional learning methods and deep user representation learning methods. Then the validity of the model is test?ed. According to the actual test results,when the BCCM model is used to process the unbalanced data set,the values of R,F1 and F2 obtained are higher than those of the traditional decision tree,logistic regression and naive bayesian model. By processing the balanced data set,it can be found that the BCCM model achieves better results than the traditional baseline classification model. Through the comprehensive analysis of the experimental results,it can be found that the in-depth user representation learning meth?od and specific text content data can be used in the research of user portrait identification of perceptual investors,which is condu?cive to the comprehensive improvement of user portrait evaluation indicators.关键词
用户画像/情感分析/用户学习/特征融合Key words
user portrait/emotional analysis/user learning/fusion characteristics分类
信息技术与安全科学引用本文复制引用
周晓华..基于深度学习BCCM模型的网上用户画像识别分析∗[J].计算机与数字工程,2019,(9):2176-2179,2226,5.基金项目
国家自然科学基金项目(编号:61725103)资助. （编号:61725103）
