基于企业社交网络的员工幸福感研究
发布时间：2018-10-09 21:03
【摘要】：2012年央视有关“幸福”的调查,将“幸福”这个千百年来人们一直不断追求的目标再次引入公众的视野,顿时在社会上引起了广泛而激烈的讨论。此外,社交网络的发展已经日渐成熟,人们能够在上面交流、交友,在工作、生活之余吐露自己的心声。本文以企业社交网络为基础,借助情感分析、文本分析及复杂网络分析等计算机技术,对员工的幸福感展开研究。主要研究了基于文本分析的幸福感测量方式,并分析了人口统计学特征和网络结构特征与幸福感及其基本情感之间的相关关系。主要研究内容及成果如下： (1)以Ren-CECps词库为基础,构建幸福感测量模型。在文献研究的基础上,本文认为员工的情感幸福感,是积极情感体验与消极情感体验所共同组成的总的情感体验。本研究摒弃了幸福感传统的自我报告式(self-reported)的测量方式,转而借助Ren-CECps词库中每个词汇所包含的8种基本情感,通过中文分词、情感分析等计算机技术,构建并实现了中文环境下幸福感的测量模型。 (2)采用设定阈值、反复试验的方式,对Ren-CECps词库进行改造。为使幸福感测量模型的测量结果,表现出更明显的波动性,本文提出采用设定阈值,并将词库中各情感分量均不超过阈值的词汇删除的方式,对词库进行进一步的筛选和改造。最终通过反复试验,得出当阈值为0.4时,能够使词库的覆盖率及幸福感的波动性达到一个较好的平衡点。 (3)结合中国某电信运营商某省分公司内部班组博客系统,对员工的幸福感进行测量,并做了员工幸福感与其人口统计学特征及网络结构特征之间的相关性分析。得出企业员工的幸福感与其在企业社交网络中的地位、位置等没有直接的相关关系。
[Abstract]:A 2012 CCTV survey on "happiness" has reintroduced happiness, a goal that people have been pursuing for thousands of years, into the public eye, and has immediately aroused widespread and heated discussion in society. In addition, the development of social networks has matured, where people can communicate, make friends, and express their feelings after work and life. Based on the enterprise social network, this paper studies the employee's happiness with the help of computer technology, such as emotion analysis, text analysis and complex network analysis. This paper mainly studies the measurement method of happiness based on text analysis, and analyzes the correlation between demographic characteristics and network structure characteristics, happiness and its basic emotions. The main research contents and results are as follows: (1) based on Ren-CECps lexicon, a well-being measurement model is constructed. On the basis of literature study, this paper holds that the emotional well-being of employees is a total emotional experience composed of positive emotional experience and negative emotional experience. This study abandoned the traditional self-reporting (self-reported) measurement of happiness, and instead used the eight basic emotions contained in each word in the Ren-CECps lexicon, using Chinese word segmentation, affective analysis and other computer techniques. The measurement model of happiness in Chinese environment is constructed and realized. (2) the Ren-CECps lexicon is modified by setting the threshold and experimenting repeatedly. In order to make the measurement results of happiness measurement model show more obvious volatility, this paper proposes a method of selecting and modifying the lexicon by setting a threshold and deleting words whose emotional components do not exceed the threshold in the thesaurus. Finally, through repeated experiments, it is concluded that when the threshold is 0.4, the lexicon coverage and the volatility of well-being can reach a good equilibrium point. (3) combined with the internal group blog system of a branch of a telecom operator in China, this paper measures the well-being of the employees, and analyzes the correlation between the well-being of the employees and the demographic characteristics and network structure characteristics of the employees. It is concluded that there is no direct correlation between the employee's happiness and their position and position in the enterprise social network.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F272.92;F224
本文编号：2260780
[Abstract]:A 2012 CCTV survey on "happiness" has reintroduced happiness, a goal that people have been pursuing for thousands of years, into the public eye, and has immediately aroused widespread and heated discussion in society. In addition, the development of social networks has matured, where people can communicate, make friends, and express their feelings after work and life. Based on the enterprise social network, this paper studies the employee's happiness with the help of computer technology, such as emotion analysis, text analysis and complex network analysis. This paper mainly studies the measurement method of happiness based on text analysis, and analyzes the correlation between demographic characteristics and network structure characteristics, happiness and its basic emotions. The main research contents and results are as follows: (1) based on Ren-CECps lexicon, a well-being measurement model is constructed. On the basis of literature study, this paper holds that the emotional well-being of employees is a total emotional experience composed of positive emotional experience and negative emotional experience. This study abandoned the traditional self-reporting (self-reported) measurement of happiness, and instead used the eight basic emotions contained in each word in the Ren-CECps lexicon, using Chinese word segmentation, affective analysis and other computer techniques. The measurement model of happiness in Chinese environment is constructed and realized. (2) the Ren-CECps lexicon is modified by setting the threshold and experimenting repeatedly. In order to make the measurement results of happiness measurement model show more obvious volatility, this paper proposes a method of selecting and modifying the lexicon by setting a threshold and deleting words whose emotional components do not exceed the threshold in the thesaurus. Finally, through repeated experiments, it is concluded that when the threshold is 0.4, the lexicon coverage and the volatility of well-being can reach a good equilibrium point. (3) combined with the internal group blog system of a branch of a telecom operator in China, this paper measures the well-being of the employees, and analyzes the correlation between the well-being of the employees and the demographic characteristics and network structure characteristics of the employees. It is concluded that there is no direct correlation between the employee's happiness and their position and position in the enterprise social network.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F272.92;F224
【参考文献】
相关期刊论文 前5条
1 朱鹏;姚亦锋;张培刚;;基于人的“需求层次”理论的“宜居城市”评价指标初探[J];河南科学;2006年01期
2 吴明霞;30年来西方关于主观幸福感的理论发展[J];心理学动态;2000年04期
3 邱林;郑雪;;主观幸福感的结构及其与人格特质的关系[J];应用心理学;2005年04期
4 郑雪,王玲,邱林,严标宾;大学生主观幸福感及其与人格特征的关系[J];中国临床心理学杂志;2003年02期
5 严标宾,郑雪,邱林;主观幸福感研究综述[J];自然辩证法通讯;2004年02期
，本文编号：2260780
本文链接：https://www.wllwen.com/jingjilunwen/jingjiguanlilunwen/2260780.html
