Personality-based refinement for sentiment classification in microblog

Using social media and personality traits to assess software developers' emotional polarity

Although human factors (e.g., cognitive functions, behaviors and skills, human error models, etc.) are key elements to improve software development productivity and quality, the role of software developers' emotions and their personality traits in software engineering still needs to be studied. A major difficulty is in assessing developers' emotions, leading to the classic problem of having difficulties understanding what cannot be easily measured. Existing approaches to infer emotions, such as facial expressions, self-assessed surveys, and biometric sensors, imply considerable intrusiveness on developers and tend to be used only during normal working periods. This article proposes to assess the feasibility of using social media posts (e.g., developers' posts on Twitter) to accurately determine the polarity of emotions of software developers over extended periods in a non-intrusive manner, allowing the identification of potentially abnormal periods of negative or positive sentiments of developers that may affect software development productivity or software quality. Our results suggested that Twitter data can serve as a valid source for accurately inferring the polarity of emotions. We evaluated 31 combinations of unsupervised lexicon-based techniques using a dataset with 79,029 public posts from Twitter from sixteen software developers, achieving a macro F1-Score of 0.745 and 76.8% of accuracy with the ensemble comprised of SentiStrength, Sentilex-PT, and LIWC2015_PT lexicons. Among other results, we found a statistically significant difference in tweets' polarities posted during working and non-working periods for 31.25% of the participants, suggesting that emotional polarity monitoring outside working hours could also be relevant. We also assessed the Big Five personality traits of the developers and preliminarily used them to ponder the polarities inferences. In this context, Openness, Conscientiousness, and Extraversion were frequently related to neutral and positive posts, while Neuroticism is associated with negative posts. Our results show that the proposed approach is accurate enough to constitute a simple and non-intrusive alternative to existing methods. Tools using this approach can be applied in real software development environments to support software team workers in making decisions to improve the software development process.

Encoding Text Information with Graph Convolutional Networks for Personality Recognition

Personality recognition is a classic and important problem in social engineering. Due to the small number and particularity of personality recognition databases, only limited research has explored convolutional neural networks for this task. In this paper, we explore the use of graph convolutional network techniques for inferring a user’s personality traits from their Facebook status updates or essay information. Since the basic five personality traits (such as openness) and their aspects (such as status information) are related to a wide range of text features, this work takes the Big Five personality model as the core of the study. We construct a single user personality graph for the corpus based on user-document relations, document-word relations, and word co-occurrence and then learn the personality graph convolutional networks (personality GCN) for the user. The parameters or the inputs of our personality GCN are initialized with a one-hot representation for users, words and documents; then, under the supervision of users and documents with known class labels, it jointly learns the embeddings for users, words, and documents. We used feature information sharing to incorporate the correlation between the five personality traits into personality recognition to perfect the personality GCN. Our experimental results on two public and authoritative benchmark datasets show that the general personality GCN without any external word embeddings or knowledge is superior to the state-of-the-art methods for personality recognition. The personality GCN method is efficient on small datasets, and the average F1-score and accuracy of personality recognition are improved by up to approximately 3.6% and 2.4–2.57%, respectively.

The Effect of User Psychology on the Content of Social Media Posts: Originality and Transitions Matter

Multiple studies suggest that frequencies of affective words in social media text are associated with the user's personality and mental health. In this study, we re-examine these associations by looking at the transition patterns of affect. We analyzed the content originality and affect polarity of 4,086 posts from 70 adult Facebook users contributed over 2 months. We studied posting behavior, including silent periods when the user does not post any content. Our results show that more extroverted participants tend to post positive content continuously and that more agreeable participants tend to avoid posting negative content. We also observe that participants with stronger depression symptoms posted more non-original content. We recommend that transitions of affect pattern derived from social media text and content originality should be considered in further studies on mental health, personality, and social media.