The ASONAM conference, which stands for Advances in Social Network Analysis and Mining, is an interdisciplinary venue that encourages and acts as a platform for international researchers, experts, and scholars interested in mining social data. This international conference combines empirical, experimental, methodological, and theoretical research on social network analysis and mining, and it is wide in scope, covering method development in graph theory, statistics, data mining, machine learning, and statistical mechanics. The conference proceedings are published by the Association for Computing Machinery (ACM) and the Institute of Electrical and Electronics Engineers (IEEE). 

This conference accepts papers based on a body of key themes—namely, a focus on specific techniques, problems, and application domains. It values techniques like statistical learning algorithms, data collection/quality, and big data and scalability. For problems, the conference values papers that solve issues in network evolution, link prediction, centrality, and community detection. Any papers with application domains relevant to network identification or characterization (e.g., semantic networks, probabilistic networks, online & offline networks, etc.) are presented at this conference. As such, this annual conference has been crucial in developing methods for data mining that can be used in the interdisciplinary field (i.e., sociology, anthropology, psychology, economics, and information and computer sciences) to study the Internet, social media, and other large-scale, socio-technological infrastructures. From the combined research of these fields, this conference has become top-tier in its applications and insights into data mining.

This year, Professor Nitin Agarwal continued to serve as the General Co-Chair for the ASONAM conference and was recognized by the conference’s Steering Chair and Organizers for his exceptional contributions as chair.

“ASONAM is one of the very few flagship conferences for the international social computing research community,” said Dr. Agarwal. He continued, “Having several of our research studies ranging from suspicious behaviors to recommender algorithm bias to multi-agent based systems to multi-platform analysis leveraging knowledge graph presented at the conference provides our students with an incredible, invaluable, and interdisciplinary experience.” 

Several cosmographers were invited to present at the ASONAM 15th International Conference, and one of the key themes explored was the role of computational approaches in understanding and combating misinformation in the digital age. The following cosmographers and their papers were accepted and presented:

  • Abiola Akinnubi, Nitin Agarwal, Mustafa Alassad, and Jeremiah Ajiboye, presenting their paper Knowledge Graph Embedding for Topical and Entity Classification in Multi-Source Social Network Data
  • Ishmam Solaiman and Nitin Agarwal, presenting their paper Multiagent-based Youtube Content Discovery Bot
  • Mert Can Cakmak, Obianuju Okeke, Ugochukwu Onyepunuka, Billy Span, and Nitin Agarwal, presenting their paper Analyzing Bias in Recommender Systems: A Comprehensive Evaluation of YouTube’s Recommendation Algorithm
  • Shadi Shajari, Mustafa Alassad, and Nitin Agarwal, presenting their paper Characterizing Suspicious Commenter Behaviors

This conference gave our cosmographers the opportunity to publish their research internationally and interact with a wide range of researchers, especially due to the fact that it was co-located with several other related conferences/events. Specifically, ASONAM 2023 was co-located with FAB 2023 (Foundations and Applications of Big Data Analytics), FOSINT-SI 2023 (Foundations of Open Source Intelligence and Security Informatics), and HI-BI-BI 2023 (Network Enabled Health Informatics, Biomedicine, and Bioinformatics). Thus, this conference introduced our cosmographers to many researchers across several fields and research applications.

The core of ASONAM resides in how immensely interdisciplinary it is, where scholars spanning a spectrum of fields can unite to delve into the intricacies of individual-level and group-level behaviors gleaned by mining social data. The interdisciplinary methodology encouraged by this conference helps researchers of non-computer science fields discover algorithmic and machine-learning data mining methods they can use across their varied research.

Abiola Akkinubi, speaking on Knowledge Graph Embedding for Topical and Entity Classification in Multi-Source Social Network Data:

“This research focuses on the challenge of utilizing knowledge graphs for multi-source social network data. We used the proposed solutions to collect data from blogs, Reddit, Twitter, and YouTube. Next, we employed our framework to create a heterogeneous knowledge graph by modeling the data from multiple platforms. We stored the modeled and extracted knowledge graph using Neo4J. We extracted various relationships across these documents and performed topic and entity classification, as well as knowledge graph embedding, to establish missing ties across entities, topics, and documents. Additionally, we compared our approach with other pre-trained knowledge embeddings that were applied to publicly available data, using four different scoring mechanisms. This work was presented at the 2023 ASONAM conference in Turkey as part of my dissertation titled ‘Modeling Knowledge Graphs And Evaluating Applicability In Multi-Source Social Network Data.’”

Shadi Shajari, speaking on Characterizing Suspicious Commenter Behaviors: 

“This study presents an approach based on social network analysis to detect suspicious commenter behaviors and identify similarities across 20 YouTube channels that disseminated false views about the US Military. To achieve this goal, a combination of methods were employed, including Graph2vec, UMAP, K-means, Hierarchical clustering, qualitative and quantitative analyses. The objective was to categorize channels based on the level of suspicious behavior and reveal common patterns exhibited among them. To assess the effectiveness of the proposed methodology, the outcomes revealed the presence of commenter mobs and significant similarities among these channels, providing valuable insights into the prevalence of suspicious commenter behavior. The presentation was conducted online through a recorded video of the presentation.”

Mert Can Cakmak, speaking on Analyzing Bias in Recommender Systems: A Comprehensive Evaluation of YouTube’s Recommendation Algorithm: 

“This research investigates potential biases in YouTube’s recommendation algorithm. It focuses on how the algorithm impacts the dissemination of content related to three socio-political narratives: the China-Uyghur crisis, Cheng Ho propaganda, and the South China Sea conflict. The study aims to identify biases, highlighting concerns about filter bubbles and echo chambers in recommender systems. The methodology involved analyzing ‘seed’ videos and their recommended (depth) videos across three narratives. The findings offer insights into bias in recommender systems, aiding the development of strategies to mitigate these biases. The research can inform policymakers and platform developers to create fairer and more diverse automated decision-making systems. This is crucial for enhancing user experience and establishing effective guidelines and policies for recommender systems.

“The talk at the conference was a success. People were really interested in the subject, showing how much everyone cares about the way online recommendation systems affect our lives. There was a lot of excitement and curiosity, with many people asking questions after the talk. This showed just how important and relevant the research is. The audience was also keen to see more work in this area, asking for further studies to better understand the influence of these systems.”