CrediBot: Applying Bot Detection for Credibility Analysis on Twitter

Abstract

Nowadays, people and organizations use social networks for allowing and facilitating the transfer of information among groups that share similar interests. Due to the wide repertoire of users that these social platforms have and the amount of information generated within them, the presence of bots has become a relevant issue, both to facilitate the sharing of true information or to disseminate false information (fake news). In the second case, bots could manipulate political opinions, be perpetrators of identity or information theft, among other possible dangers that can cause when interacting on the platform. Thus, the identification of bots in social networks can become a useful practice to evaluate credibility or to detect fake news. In this work, we extend a previously proposed credibility model for Twitter, by incorporating bot detection. The original model calculates the credibility of tweets based on three measures: text, account/user, and social impact, using different filters to analyse text (SPAM, bad words,and good spelling) and account attributes (e.g., creation date, followers, following) to calculate account/user and social credibility scores. The extended model considers in the user credibility, the bot verification. Additionally, the extended credibility model is implemented in T-CREo, a framework for real time credibility analysis. For bot detection, some machine learning algorithms for supervised learning, such as AdaBoost, Bagging, Decision Tree, Logistic Regression, and Random Forest are trained and evaluated. Results show that the best algorithm is the Random Forest for its capacity of generalization with an accuracy and F1-score values over 97% both in English and Spanish. The evaluation of the bot detection functionality in the credibility analysis shows a performance of precision=1.0, recall=0.8462, F1-score=0.9167, and accuracy=0.92 for both English and Spanish models in our validation tests.

MSCMGTB: A Novel Approach for Multimodal Social Media Content Moderation Using Hybrid Graph Theory and Bio-Inspired Optimization

“…The precision levels based anon these assessments are displayed as follows in Figure 4 SGNN [8] CrediBot [12] ALDA [17] MSCMGTB At the outset, it's evident that the precision rates fluctuate across different NTS values for all models. For instance, when analyzing smaller datasets (84k NTS), SGNN shows a precision rate of 79.04%, while MSCMGTB scores slightly higher at 86.78%.…”

“…These studies highlight the diverse applications of machine learning in understanding social media contents. Aguilera et al [12] and Guo et al [13] contributed to the field by applying bot detection for credibility analysis on Twitter and mitigating the influence of disinformation propagation, respectively for different scenarios. These studies underscore the importance of credibility and integrity in online contenst.…”

“…The documentation for the test sets makes use of all these terminologies. To determine the appropriate TP, TN, FP, and FN values for these scenarios, we compared the projected Harmful Social Media Instances likelihood to the actual Harmful Social Media Instances status in the test dataset samples using the Social Graph Neural Network (SGNN) [8], CrediBot [12], and Adaptive LDA (ALDA) [17] techniques.…”

“…The integration of CNNs for image analysis and Transformers for textual analysis, combined with the bi-directional attention mechanism, enables MSCMGTB to effectively process and understand the complex interplay between different types of content. This multi-faceted approach allows for a more SGNN [8] CrediBot [12] ALDA [17] MSCMGTB nuanced understanding of social media content, leading to higher accuracy in identifying inappropriate material.…”

“…This indicates MSCMGTB's strong ability to correctly identify non-harmful content, reducing the likelihood of falsely flagging benign content as harmful. SGNN [8] CrediBot [12] ALDA [17] MSCMGTB However, there are instances where MSCMGTB's specificity fluctuates. For instance, at 264k NTS, its specificity is 82.62%, which, while competitive, does not stand out as distinctly superior to the other models.…”