Analysis Of Emotional Expression Of Weibo Users With Different Personality Disposition

As the largest social network media in china.Weibo has become an important platform for netizens to express their opinions and emotion.The emotional information contained in the social media media will have an impact on others and may play a non-negligible role in the dissemination of public events,which makes it necessary to explore the individual's emotional expression characteristics in such environments.Personality is an important factor affecting emotional expression.In the past,when studying the influence of personality on emotional expression,stress scenarios were used as experimental context,and emotional expression was regarded as an emotional coping strategy when individuals faced stress.For the social media environment,the influence of personality on emotional expression has been studied less.Therefore,this study analyze the emotion expressional characteristics of different personality-oriented users in weibo platform as environments.Because it takes time and effort to obtain a large number of Weibo users' personality scores by using traditional questionnaires,this study chooses to establish a classification model to achieve the personality prediction of Weibo users,and then uses online textual analysis technology to analyze the emotional expression characteristics of different personality types of Weibo usersIn study 1,the researcher take the expressive vocabularies that are most relevant to the dimensions of the Big Five personality in the previous research as the seed words for translation and classification.Then randomly crawling the original microblog of 3374 Chinese active Weibo users as a corpus.Using Word2Vec and the corpus to expand the translated seed words:based on the principle of consensus,two psychology graduate students screened the expanded vocabularies to form the first version of the personality lexicon;Then,three classification machine learning algorithms are selected to verify the first version of personality lexicon,and the first version of the personality lexicon is screened and revised according to the feature importance in the random forest results to form the final version of the personality lexicon and prediction model.In study 2.according to the personality lexicon and prediction model obtained in Study 1,1868 Weibo users who were randomly crawled from the Weibo platform were labeled with personality tags,divided into high and low disposition.and then according to the existing emotion lexicon,the negative lexicon and vocabulary matching technique,the sentiment analysis was carried out to calculate the proportion of Weibo entries containing each emotional category of emotional words,and use Bavesian t test to analyze the results of the two personality-oriented users.Finally,The results of the sentiment anal\ysis of high-bias users in each dinmension of the Big Five personality are summarizedIn sum.the present study drew the following coclusions1)In mixed emotions,users higher in openness will express less negative emotion:users higher in conscientiousness will express less negative emotion;extrovert users will express more positi-ve emotion and expression is more emotional;users higher in a(greeableness users will express more positive emotion,less negative emotion;high neuroticism users will express more negative emotion and expression is more emotional.2)In the 8 basic emotions,users higher in openness will express less ansger;users higher in conscientiousness will express less anger;extrovert users will express more happiness and expression is more emotional;users higher in agreeableness users will express more happiness,like and expectation,express less anger,anxiety and disgust;high neuroticism users will express more anger,anxiety,disgust and sadness.