As society moves toward digitization, various informal learning environments have emerged, such as social media. However, it remains to be studied whether adolescents who are active in commenting on social media achieve higher stages of knowledge construction. This study adopts a mixed research method, combining random sampling, automated coding, social network analysis, regression, and others to explore the relationship between learner comments and their knowledge construction in the social media platform. The results show that learners' closeness centrality and betweenness centrality are significant predictors of the stage of knowledge construction, with the former having a positive effect and the latter a negative effect. The number of characters in comments, speeches in comments, and likes on comments also significantly predict the stage of knowledge construction, but with smaller effects. Additionally, this study validates the accuracy of automated coding by generative artificial intelligence (GAI) through comparisons with machine learning models such as Bert.