Research Nuggets: Cyber Safety/Security

In these projects we explore several aspects of cyber safety and security. For example, social media and chat rooms have exploded in popularity in recent years. Social Media pervades much of our daily life and has significant socio-economic impact around the world. Several interesting research issues arise in this context. Can you trust the information you read in these media? Is it true or fake? Who spread that rumor? Can you trust the gender or identity of the person you are conversing with on social media? Is the person you are communicating with intentionally trying to deceive you? What is the nature of information propagation through this medium? How does information spread virally? What are the human dynamics in Social Media and how does it impact information consumption? These are just some of the questions we explore in the research direction. Our work has resulted in several papers, patents and tools that have been used widely on the Internet. Below are some of the more recent research nuggets in this area.

interpretable architecture for fake news detection

Mingxuan Chen, Ning Wang and K. P. Subbalakshmi, Explainable Rumor Detection using Inter and Intra-feature Attention Networks,TrueFact Workshop, KDD 2020

One of the serious problems that has emerged with proliferation of social media, is the propagation of rumors or fake news or misinformation. Therefore, rumor identification is a very critical task with significant implications to economy, democracy as well as public health and safety. We tackle the problem of automated detection of rumors in social media in this work by designing a modular explainable architecture that uses both latent and handcrafted features and can be expanded to as many new classes of features as desired. This approach allows the end user to not only determine whether the piece of information on the social media is real of a rumor, but also give explanations on why the algorithm arrived at its conclusion. Using attention mechanisms, we are able to interpret the relative importance of each of these features as well as the relative importance of the feature classes themselves. The advantage of this approach is that the architecture is expandable to more handcrafted features as they become available and also to conduct extensive testing to determine the relative influences of these features in the final decision. Extensive experimentation on popular datasets and benchmarking against eleven contemporary algorithms, show that our approach performs significantly better in terms of F-score and accuracy while also being interpretable.



Who Spread that Rumor?

Alireza Louni and K.P. Subbalakshmi, ``Who Spread that Rumor: Finding the Source of Information in Large On-line Social Networks with Probabilistically Varying Inter-Node Relationship Strengths", IEEE Transactions on Computational Social Systems, Feb 2018.

We address the problem of estimating the source of a rumor in large-scale social networks. Previous works studying this problem have mainly focused on graph models with deterministic and homogenous internode relationship strengths. However, internode relationship strengths in real social networks are random. We model this uncertainty by using random, non-homogenous edge weights on the underlying social network graph. We propose a novel two-stage algorithm that uses the modularity of the social network to locate the source of the rumor with fewer sensor nodes than other existing algorithms. We also propose a novel method to select these sensor nodes. We evaluate our algorithm using a large data set from Twitter and Sina Weibo. Real-world time series data are used to model the uncertainty in social relationship strengths. Simulations show that the proposed algorithm can determine the actual source within two hops, 69%-80% of the time, when the diameter of the networks varies between 7 and 13. Our numerical results also show that it is easier to estimate the source of a rumor when the source has higher betweenness centrality. Finally, we demonstrate that our two-stage algorithm outperforms the alternative algorithm in terms of the accuracy of localizing the source.

Related patent and publications

  • Mingxuan Chen, Xinqiao Chu, K. P. Subbalakshmi, "MMCoVaR: Multimodal COVID-19 Vaccine Focused Data Repository for Fake News Detection and a Baseline Architecture for Classification", ASONAM 2021. [download]

  • Mingxuan Chen, Ning Wang, K. P. Subbalakshmi, ``Explainable Rumor Detection using Inter and Intra-feature Attention Networks", True Fact Workshop KDD, August 2020.

  • Alireza Louni and K.P. Subbalakshmi, ``Who Spread that Rumor: Finding the Source of Information in Large On-line Social Networks with Probabilistically Varying Inter-Node Relationship Strengths", IEEE Transactions on Computational Social Systems, Accepted for Publication, 2018.

  • Alireza Louni and K.P. Subbalakshmi, "Method and Apparatus to Indentify the Source of Information/Misinformation in Large Scale Networks", Provisional Patent Filed, January 2015

  • Alireza Louni, Anand Santhanakrishnan and K.P. Subbalakshmi. (Aug 19, 2015). "Identification of Source of Rumors in Social Networks with Incomplete Information", 2015 ASE Eighth International Conference on Social Computing (SocialCom 2015). Academy of Science and Engineering. (9% Acceptance Rate). Download (535 kb PDF).

  • Alireza Louni and K.P. Subbalakshmi, A Two-stage Algorithm to Estimate the Source of Information Diffusion in Social Media Networks", IEEE INFOCOM Workshop on Dynamic Social Networks, April 2014

  • S. Anand, M. Venkataraman, K. P. Subbalakshmi and R. Chandramouli. (Oct 1, 2015). "Spatio-Temporal Analysis of Passive Consumption in Internet Media", IEEE Transactions on Knowledge and Data Engineering, IEEE. 27 (10), 2839-2850.

  • Alireza Louni and K.P. Subbalakshmi. (2014). "Diffusion of Information in Social Networks", Social Networking: Mining, Visualization and Security, Mrutyunjaya Panda, Satchidananda Dehuri and Gi-Nam Wang, Springer-Verlag GmbH. XXII.

  • S. Anand, K.P. Subbalakshmi and R. Chandramouli. (Jan 2013). "A Quantitative Model and Analysis of Information Confusion in Social Networks", IEEE Transactions on Multimedia, 15 (1), 207 - 223. Download (375 kb PDF).

  • M. Venkataraman, K.P. Subbalakshmi and R. Chandramouli. (May 2012). "Measuring and quantifying the silent majority on the Internet", IEEE Sarnoff Symposium. IEEE. Invited Paper.

  • S. Anand, R. Chandramouli, K.P. Subbalakshmi and M. Venkataraman,"Altruism in social networks: Good guys do finish first", Springer Social Networks Analysis and Mining, Print ISBN:1869-5450, Mar 2012.

  • S. Anand, R. Chandramouli and K.P. Subbalakshmi, "Cost of collaboration vs. individual effort in social networks", SIAM Data Mining Conference, 2011.

  • Xiaoling Chen, R. Chandramouli and K.P. Subbalakshmi. (2014). "Scam Detection in Twitter", Data Mining for Service, Katsutoshi Yada, Series: Studies in Big Data, Springer Berlin Heidelberg. 3 133-150.

  • Z. Dong, R.D.W. Perera, R. Chandramouli and K.P. Subbalakshmi. (Jan 2012). "Network measurement based modeling and optimization for IP geo-location", Elsevier Journal Computer Networks: The International Journal of Computer and Telecommunications Networking, 56 (1), 85-98.

  • X. Chen, R. Chandramouli and K.P. Subbalakshmi. (2011). "Scam detection in Twitter", SIAM Text Mining Workshop . SIAM.

  • Na Cheng, R. Chandramouli and K.P. Subbalakshmi. (2011). "Author Gender Identification from Text Documents", The Journal of Digital Investigation Volume 8, Issue 1, July 2011, Pages 78–88.

  • Xiaoling Chen, Rohan D Perera, Ziqian Dong, R. Chandramouli, and K.P. Subbalakshmi. (2010). "Deception Detection on the Internet", Research on Computational Forensics, Digital Crime and Investigation: Methods and Solutions, IGI Global, ISBN13: 978-1-60566-836-9. 334 – 354.

  • X. Chen, R. Chandramouli and K.P. Subbalakshmi. (2011). "Authorship similarity detection from emails", Machine Learning and Data Mining Conference.

  • P. Hao, X. Chen, R. Chandramouli and K.P. Subbalakshmi. (2011). "Adaptive context modeling for deception detection in emails", Machine Learning and Data Mining Conference.

  • Na Cheng, Xiaoling Chen, Rajarathnam Chandramouli and K.P. Subbalakshmi. (2008). "Gender Identification from E-mails", IEEE Symposium on Computational Intelligence and Data Mining.