Comment sections of an online post enable social media users to express their ideas and form virtual discussions. Most studies focused only on user-to-user relationships and ignoring the valuable information from those digital conversations, which is essential for understanding online communication behavior. This study proposes a novel approach to retrieve popular patterns on online conversations using network-based analysis. The analysis consists of two main stages: intent analysis and network generation. Users’ intention is detected using keyword-based classification, followed by the utilization of Naïve Bayes and Support Vector Machines algorithms for uncategorized comments. Afterward, human-in-the-loop is involved in improving the keyword-based classification. To extract essential information on social media collection and communication patterns among the users, we build conversation graphs using a direct multigraph network. The experiments categorize 90% comments with 98% accuracy for a real social media challenge data: YourExpo2015. Finally, the most popular online discussion patterns obtained from conversation graphs resemble real-life communication activities.