Quantifying Chinese Happiness via Large-Scale Microblogging Data

Happiness is an important indicator to measure our life satisfaction. Microblogging data can reflect users' living standards and psychological state. We build a large lexicon based on the PERMA lexicon and large-scale microblogging data on Sina Weibo by combining the PMI and distrabutional similarty method. Using this lexicon, we propose a method to calculate the happiness quantitatively based on PERMA theory. Experiments shows that our method achieve significant improvement compared with the baseline in terms of AP and Bpref respectively. After the verification of our method on manual annotation dataset, we perform an in-depth analysis on large-scale microblogging data using our method.