提示
基于boosting的幸福感预测
首发时间：2021-01-20
摘要：当今中国社会，幸福已得到了越来越多的关注。但是，从大数据的角度探讨幸福的研究还不多。尽管在心理学研究中，大数据方法的应用并不广泛，但机器学习、深度学习等大数据技术，为传统心理学研究提供了全新的研究方法，带来了新的思路。本文基于中国人民大学中国调查与数据中心主持的《中国综合社会调查（cgss）》项目的数据，通过使用数据挖掘的技术来进行幸福感预测的研究，探究幸福感的影响因素和幸福感预测模型的效果。研究按照因子分析、特征工程、模型建立和评估三个阶段，系统地利用数据科学的方法进行幸福感影响因子的分析和幸福感预测模型的设计。研究发现：社会态度（公平）、家庭变量（家庭资本）、个体变量（心理健康、社会经济地位、社会等级）对幸福感的影响较大，且在利用这五个特征建立的幸福感预测模型中，基于boosting算法的模型和模型融合的效果较优。
关键词：
for information in english, please click here
the prediction of happiness based on boosting
abstract：in today\'s chinese society, happiness has attracted more and more attention. however, there are not many studies on happiness from the perspective of big data. although in psychology research, the application of big data methods is not widespread, or even just started, but big data technologies such as machine learning and deep learning provide new research methods for traditional psychology research and bring new ideas. this article is based on the data of the "china comprehensive social survey (cgss)" project hosted by the china survey and data center of renmin university of china. it uses data mining technology to conduct research on happiness prediction and explore the influencing factors of happiness and the effect of the prediction models. according to the three stages of factor analysis, feature engineering, model establishment and evaluation, the research systematically uses data science methods to analyze the factors affecting happiness and design the prediction model of happiness. the study found that: social attitudes (fairness), family variables (family capital), and individual variables (mental health, socioeconomic status, and social rank) have a greater impact on happiness, and among the happiness prediction models established by these five features, the effect of the model based on boosting algorithm and model fusion is better.
keywords：
基金：
论文图表：
引用
导出参考文献
no.****
动态公开评议
共计0人参与
勘误表
基于boosting的幸福感预测
评论
全部评论