Over millions of years, the vertebrate immune system has evolved into one of the most complex and intelligent biological systems. The immune system's function is to protect the body from harmful intruders. Several mathematical models have been proposed to understand the adaptive immune system and its functional subsystems. We develop a novel agent-based model of T-Cell cross-regulation in the adaptive immune system, and we apply it to binary classification problems analogous to those faced by the immune system.

We expect our study to help immunologists better understand the general mechanism behind T-cell cross-regulation and also to raise questions about the behavior of the adaptive immune system in general such as immune memory, cell death and homeostasis. However, our chief aim is to show that cross-regulation dynamics can be used to classify textual documents in changing corpora. We validate the model on real-world data from biomedical articles and personal e-mails, and we compare our algorithm with other machine learning classifiers. Finally, we discuss how the guiding of T-cell self-organizing dynamics can be seen as a general system of classification, the study of which is helpful for complex systems, text classification, and theoretical immunology.