Conversation Graphs in Online Social Media

Abstract

In online social media platforms, users can express their ideas by posting original content or by adding comments and responses to existing posts, thus generating virtual discussions and conversations. Studying these conversations is essential for understanding the online communication behavior of users. 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 categorization of posts and comments, integrated with classification through Naïve Bayes and Support Vector Machine algorithms for uncategorized comments. A continuous human-in-the-loop approach further improves the keyword-based classification. To build and understand communication patterns among the users, we build conversation graphs starting from the hierarchical structure of posts and comments, using a directed multigraph network. The experiments categorize 90% comments with 98% accuracy on a real social media dataset. The model then identifies relevant patterns in terms of shape and content; and finally determines the relevance and frequency of the patterns. Results show that the most popular online discussion patterns obtained from conversation graphs resemble real-life interactions and communication.

Analyzing Brand Awareness Strategies on Social Media in the Luxury Market: the Case of Italian Fashion

“…The proposed methodology can be used by brands to improve their marketing strategies on social media. As a potential implication resulting from the proposed methodology, we propose analyzing the social conversation graphs, as designed by Brambilla et al [49,50] to build conversational agents [52,53] that potentially elevate the users' engagement.…”

“…Studying these events provides useful insights for the brands for future decisionmaking purposes. Motivated by the LRLE's potential as discussed in various works [45][46][47][48][49][50][51], we chose Milano Fashion Week, which is one of the Big Four fashion capitals.…”

A multi-perspective approach for analyzing long-running live events on social media. A case study on the “Big Four” international fashion weeks

“…Future work concerns applying the proposed framework on other LRLEs from different context such as Comic-Con to evaluate the effectiveness of the framework. Moreover, motivated by the study on conversation graphs in online social media [16], we are designing conversation agents capable of participating in some discussions [93] during LRLEs. Such conversation agents would be beneficial for the LRLEs organizers to facilitate the customer relationship management [36,78].…”

“…We considered the higher-level image related attributes in the same category of high-level features. It should be noted that, Dominant color and high-level features have been extracted using Microsoft Azure's Computer Vision services 16 .…”

Modeling and Moderation of COVID-19 Social Network Chat

“…For example, Dutta and Das [13] trained a support vector machine (SVM) model to detect if pairs of Facebook comments were consecutive messages from the same conversation. Similarly, SVMs have been used in conjunction with Naïve Bayes classifiers by Brambilla et al [14] in order to determine the latent meaning behind OSN comments, as well as the intentions of their authors. Paul and Gokhale [10] trained multiple supervised machine learning models (SVM, random forest, gradient boosting, multilayer perceptron and long short-term memory (LSTM) neural networks) to automatically differentiate between pro and anti-vaccination posts on Twitter.…”