A Novel Adaptive Conditional Probability-Based Predicting Model for User’s Personality Traits

摘要: With the pervasive increase in social media use, explosion of users’ generated data provides a potentially very rich source information, which plays an important role helping online researchers understand user’s behaviors deeply. Since personality traits are driving force behaviors, hence, this paper, along with network features, we first extract linguistic emotional statistical and topic features from Facebook status updates, followed by quantifying importance via Kendall correlation coefficient. And then, on basis weighted dynamic updated thresholds traits, deploy novel adaptive conditional probability-based predicting model considers prior knowledge correlations between to predict Big Five traits. In experimental work, explore existence better theoretical support for our proposed method. Moreover, same dataset, compared other methods, method can achieve -measure 80.6% when taking into account there is impressive improvement 5.8% over approaches.