The purpose of this paper is to explore the Values in Action (VIA) classification of human strengths and virtues by using unsupervised machine learning techniques, specifically topic modeling algorithms, on a sample of X (formerly known as Twitter) posts. This study aims to investigate if and to what extent the structure of posts with the highest positive sentiment, as determined by topic modeling algorithms, aligns with the structure of the VIA classification.
This study uses a sample of X posts as the data set for the analysis. Unsupervised machine learning techniques, specifically topic modeling algorithms, are used to extract and categorize topics from X posts. The sentiment analysis algorithm is used to identify posts with the most positive sentiment. The structure and representation of these positive sentiment posts are then compared with the structure of the VIA classification.
The results of this study reveal a correlation between the structure of posts with the highest positive sentiment, as determined by topic modeling algorithms, and the structure of the VIA classification. This indicates that the topic structures derived from the X posts exhibit similarities to the categorization of character strengths proposed by the VIA classification. The findings of this study provide empirical validation for the VIA classification framework when applied to social media data.
This paper contributes to the literature by using unsupervised machine learning techniques to validate the VIA classification on social media data. The use of these innovative methods adds a novel dimension to the research on character strengths and virtues.
