| 基于微博用户的情绪变化分析 |
| |
| 作者姓名： | 汪静莹 甘硕秋 赵楠 刘天俐 朱廷劭 |
| |
| 作者单位： | 1. 中国科学院心理研究所中国科学院行为科学重点实验室, 北京 100101;
2. 中国科学院大学, 北京 100049;
3. 北京大学人口研究所, 北京 100871 |
| |
| 摘 要： | 通过网络大数据的方法宏观地分析微博用户在不同季节和时间的情绪变化.以195万微博活跃用户为样本，在每个季节下载一周的微博活跃用户数据，利用“中文心理分析系统”计算每个季节积极情绪和消极情绪词的词频.结果显示：1）人们的综合情绪的2个高峰分别在中午和晚8点；2）虽然人们在周末的积极情绪与工作日无异，但消极情绪在周末明显低于工作日；3）人们在夏季的积极情绪和消极情绪最高，在秋季的积极情绪和消极情绪最低；4）两种性别的情绪走势一致，但女性较男性有更多情绪表达、更情绪化，更易感.
|
| 关 键 词： | 微博 情绪 生物节律 大数据 性别差异 |
| 收稿时间： | 2016-01-29 |
Chinese mood variation analysis based on Sina Weibo |
| |
| Authors: | WANG Jingying GAN Shuoqiu ZHAO Nan LIU Tianli ZHU Tingshao |
| |
| Institution: | 1. Key Laboratory of Behavioral Sciences, Institute of Psychology, Chinese Academy of Sciences, Beijing 100101, China;
2. University of Chinese Academy of Sciences, Beijing 100049, China;
3. Institute of Population Research, Peking University, Beijing 100871, China |
| |
| Abstract: | This study tried to recognize the Chinese mood variations in different seasons and time periods via the big data on Internet. Participants were 1.95 million active users of Sina Weibo. We downloaded their data of one week selected in each of the four seasons of one year. Then we used TextMind to calculate the ratios of positive affect (PA) and negative affect (NA) words. Results are given as follows. 1) Integrated mood had two peaks at noon and 8 o'clock in the evening. 2) Although PA in weekends was not different from that on the weekdays, NA in weekends was lower than that on weekdays. 3) In summer, both PA and NA were the highest. In autumn, both PA and NA were the lowest. 4) The two genders had similar mood variation trends, but females had more mood expressions than males and they were sentimental and susceptive. |
| |
| Keywords: | microblog mood circadian rhythm big data sex differences |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《》浏览原始摘要信息 |
| 点击此处可从《》下载免费的PDF全文 |
|
