Newly-Coined Words and Emoticon Polarity for Social Emotional Opinion Decision

Abstract

Nowadays, based-on mobile devices and internet, social network services(SNS) are common trends to everyone. Thus, decision of social and public opinions, and polarity about social happenings, political issues, government policies and decision, or commercial products is very important to the government, company, and a person. But, SNS are basically making newly-coined words and emoticons. Especially, emoticons are made by a person or companies. Newly-coined words are mostly made by communities. The SNS big data mainly consists of his kinds of newly-coined words and emoticons so that newly-coined words and emoticons analysis are very important to understand the social and public opinions, and polarity about social happenings, political issues, government policies and decision, or commercial products. Social big data is unstructured data and contains many newly-coined words and various emoticons. Therefore, there is a limitation to guarantee the accuracy and analysis range of social data of emotional analysis. The newly-coined words contains the social phenomena and trends of modern society implicitly. And the emoticons are electronic quasi-languages made up of letters and symbols, and express the emotional state more implicitly. Although the newly-coined words and emoticons are an important part of the emotional analysis, they are excluded from the emotional dictionary and analysis. In this research, newly-coined words and emoticons extracted from the raw twit messages include polarity and weight with pre-built dictionary. The polarity and weight would be calculated for emotional classification. The proposed emotional classification equation adds up the weights among the same polarity(positive or negative) and sums the negative weight value with the positive weight values. The polarity summation result is recorded in the variable. If the polarity summation result is more than threshold value, the twit message is decided as positive. If it is less than threshold value, it is decided as negative and the other values are decided as neutral. The accuracy of social big data analysis is improved by quantifying and analyzing emoticons and new-coined words.