政治網路口碑的情感分析：語意關連性之觀點

政治網路口碑勃興，其內容究竟是反映輿論，還是企圖影響輿論，一直沒有定論。本文針對2014 年臺北市長候選人連勝文與柯文哲的政治網路口碑，採情感分析詞彙法中的語料基礎取徑，以點式共通資訊（兩個詞實際上共同出現的機率與機遇的比值，越高代表越常共同出現）分別選出與連勝文或柯文哲語意關連性高的情緒詞，並檢視與人們的認知關連性是否吻合。結果發現，政治網路口碑與人們的認知不一致，受到政黨傾向與情緒字價性所調節：對負面情緒詞，無論泛藍或泛綠會將大部分歸類為與競爭對手有關；對正面的情緒詞，泛藍歸類偏向中性，泛綠則會全部歸類為與自己支持的候選人有關。結論指出政治網路口碑不是意見、沒有反映輿論，而是說服、企圖影響輿論。

關鍵字

情感分析；點式共通資訊；語意關連性；語意指向；情緒價性

並列摘要

Political word-of-mouth burgeons on the internet. Whether its content is reflecting public opinion, or actually attempting to influence public opinion, is still unknown. This study use the lexicon-based sentiment analysis to examine political word-of-mouth of the two candidates of Taipei Mayor in 2014, Sheng-Wen Lian and Wen-Zhe Ke. We first computed pointwise mutual information (PMI, the ratio of joint probability of the co-occurrence of two words to the chance of the co-occurrence of the same words) to estimate the semantic association between candidate names and emotional words. Then semantic orientation, which is the difference between the PMI of an emotional word to the two candidate names, was obtained to classify emotions as either in the Lian group or the Ke group. Finally, an online survey was used to assess whether the pattern appearing in political word-of-mouth matches the cognitive association in people’s mind. The results show that semantic orientation in political word-of-moth and cognitive association in people’s mind are inconsistent. The partisanship and the valence of emotional words moderate the aforementioned relationship. It is concluded that political word-of-mouth on the internet is in fact persuasive text, not opinion.

並列關鍵字

sentiment analysis ； pointwise mutual information ； semantic association ； semantic orientation ； valence

參考文獻

Archak, N., Ghose, A., & Ipeirotis, P. G. (2011). Deriving the pricing power of product features by mining consumer reviews. Management Science, 57, 1485-1509.doi:10.1287/mnsc.1110.1370

Boiy, E., & Moens, M.-F. (2009). A machine learning approach to sentiment analysis in multilingual Web texts. Information Retrieval, 12, 526-558. doi:10.1007/s10791-008-9070-z

Bollen, J., Mao, H., & Zeng, X. (2011). Twitter mood predicts the stock market. Journal of Computational Science, 2, 1-8. doi: 10.1016/j.jocs.2010.12.007

Brader, T. (2005). Striking a responsive chord: How political ads motivate and persuade voters by appealing to emotions. American Journal of Political Science, 49, 388-405. doi:10.1111/j.0092-5853.2005.00130.x

-- (1994). Measuring emotion: The Self-Assessment Manikin and the semantic differential. Journal of Behavioral Therapy and Experimental Psychiatry, 25, 49-59. doi:10.1016/0005-7916(94)90063-9

國際替代計量

政治網路口碑的情感分析：語意關連性之觀點

全文下載

主題瀏覽