The Agent-Based T-Cell Cross-regulation Model for Document Classification.
We have developed a bio-inspired solution for binary classification of textual documents inspired by T-cell cross-regulation in the vertebrate adaptive immune system, which is a complex adaptive system of millions of cells interacting to distinguish between self and nonself substances. In analogy, automatic document classification assumes that the interaction and co-occurrence of thousands of words in text can be used to identify conceptually-related classes of documents—at a minimum, two classes with relevant and irrelevant documents for a given concept (e.g. articles with protein-protein interaction information). Our agent-based method for document classification expands the analytical model of Carneiro et al, by allowing us to deal simultaneously with many distinct populations of antigen-specific T-Cells and their collective dynamics. We have extended this model to produce a spam-detection system. We have also developed our agent-based model further to apply it to biomedical article classification, testing it on a dataset of biomedical articles provided by the BioCreative 2.5 challenge. Our results are useful for biomedical text mining, but they also help us understand T-cell cross-regulation as a potential general principle of classification available to collectives of molecules without a central controller. While there is still much to know about the specifics of T-cell cross-regulation in adaptive immunity, Artificial Life allows us to explore alternative emergent classification principles while producing useful bio-inspired tools. Recently, we started expanding this algorithm to other forms of classification such as sensor data from human-robot interactions under an IUCRG project.
Project Members
Luis Rocha
Al Abi-Haidar
Ian Wood
Funding
Project partially funded by:
- Indiana University Collaborative Research Grants 2013. Project title: “Social SLAM: Creating Dynamical Socio-Environmental Models for Mobile Robots”.
- IARPA Contract: Early Model-Based Event Recognition with Surrogates (EMBERS), 2012-2014.
Selected Project Publications
- A. Abi-Haidar [2011]. “An adaptive document classifier inspired by T-Cell cross-regulation in the immune system” (pdf). PhD Dissertation, Indiana University
- A. Abi-Haidar and L.M. Rocha [2011]. “Collective Classification of Textual Documents by Guided Self-Organization in T-Cell Cross-Regulation Dynamics“. Evolutionary Intelligence. 4(2):69-80. DOI: 10.1007/s12065-011-0052-5.
- A. Abi-Haidar and L.M. Rocha [2010]. “Collective Classification of Biomedical Articles using T-Cell Cross-regulation“. In: Artificial Life XII: Twelfth International Conference on the Simulation and Synthesis of Living Systems. H. Fellermann et al et al (Eds.). MIT Press, pp. 706-713.
- A. Abi-Haidar and L.M. Rocha [2010]. “Biomedical Article Classification Using an Agent-Based Model of T-Cell Cross-Regulation“. In: Artificial Immune Systems: 9th International Conference, (ICARIS 2010). E. Hart, C. McEwan, J. Timmis, and A. Hone (Eds.) Lecture Notes in Computer Science. Springer-Verlag, 6209: 237-249. Recipient of Best Paper Award. for ICARIS 2010
- A. Abi-Haidar and L.M. Rocha [2008]. Adaptive Spam Detection Inspired by a Cross-Regulation Model of Immune Dynamics: A Study of Concept Drift“. In: Artificial Immune Systems: 7th International Conference, (ICARIS 2008). Bentley, Peter; Lee, Doheon; Jung, Sungwon (Eds.) Lecture Notes in Computer Science. Springer-Verlag, 5132: 36-47.
- A. Abi-Haidar and L.M. Rocha [2008]. Adaptive Spam Detection Inspired by the Immune System“. In: Artificial Life XI: Eleventh International Conference on the Simulation and Synthesis of Living Systems. S. Bullock, J. Noble, R. A. Watson, and M. A. Bedau (Eds.). MIT Press, pp. 1-8.
The Adaptive Behavior and Cognition Lab–West (ABC-West) is dedicated to exploring the cognitive mechanisms that people (and other animals) use to behave adaptively in their environments. We study the interactions between behavior and environment at multiple scales–including how cognitive mechanisms have evolved in response to particular environmental structures, how behaviors are learned through interactions with the environment, and how behaving and acting in the world can change the environmental structures that agents face in the future. We look at particular adaptively important domains such as mate choice and food choice, and we use tools including agent-based modeling and simulation and laboratory experiments. ABC-West was formed in 2005 through budding from the original 
Some of the world’s leading network scientists will soon share their latest research at a conference hosted by Indiana University.