Exploiting Parallelism Inherent in AIRS, an Artificial Immune Classifier

Abstract

The mammalian immune system is a highly complex, inherently parallel, distributed system. The field of Artificial Immune Systems (AIS) has developed a wide variety of algorithms inspired by the immune system, few of which appear to capitalize on the parallel nature of the system from which inspiration was taken. The work in this paper presents the first steps at realizing a parallel artificial immune system for classification. A simple parallel version of the classification algorithm Artificial Immune Recognition System (AIRS) is presented. Initial results indicate that a decrease in overall runtime can be achieved through fairly naïve techniques. The need for more theoretical models of the behavior of the algorithm is discussed.

Artificial Immune Network: Classification on Heterogeneous Data

“…For a consistent condition and comparison on FINERS and FAIRS (Puteh et al, 2008) and other immune algorithms from WEKA toolbox (Witten & Frank, 2005), they are tested using the same sets of 10-fold CV data. The selected immune classifiers from WEKA toolbox are AIRS1 (Watkins, 2001;, AIRS2 (Watkins & Timmis, 2002;, AIRS2Parallel (AIRS2P) Brwonlee, 2005), IMMUNOS1 Carter, 2000) and CLONALG de Castro & Von Zuben, 2000). The average accuracy is calculated from the 10 sets for each dataset and the significant difference is analyzed using paired T-Test using standard statistical package.…”

“…However, the former model uses significantly different and complex approach. The later model is the first straightforward immune-inspired supervised learning algorithm and has subsequently gone through a period of study and refinements (Watkins & Timmis, 2002;Hamaker & Boggess, 2004). However, many of these studied classification models concentrate on the population-based or clonal selection algorithm and ignore the important network feature (Timmis, 2001) of the immune system.…”

Weka Machine Learning for Predicting the Phospholipidosis Inducing Potential

“…Finally, the memory cells are used to classify novel antigens, i.e., chemical compounds in SAR studies. The structureactivity models presented here were obtained with AIRS2, an improved version of AIRS [111], as implemented by Brownlee [112]. The AIRS2 algorithm consists of the following steps:…”

WAIRS: improving classification accuracy by weighting attributes in the AIRS classifier

“…AIRS is probably the best known AIS for classification, having been developed in 2001 [1]. It has undergone a number of revisions and refinements in order to increase efficiency and increase accuracy [2,3] and has been shown to work competitively on benchmark tests using standard public domain datasets [4,5]. Indeed, when AIRS was tested on the well known Iris dataset, Pima Indians dataset, ionosphere dataset and Sonar dataset, AIRS was comparable with the fifth to eighth most successful classifiers found in the literature for three out of the four datasets (where the Pima Indians dataset was the exception) [6].…”