Biometric Tendency Recognition and Classification System: An Artistic Approach Carlos Castellanos School of Interactive Arts and Technology Simon Fraser University Surrey, B.C., Canada cca73@sfu.ca Philippe Pasquier School of Interactive Arts and Technology Simon Fraser University Surrey, B.C., Canada pasquier@sfu.ca Luther Thie Interaction Design Institute Ivrea, Italy concept@lutherthie.com Kyu Che Symbionic Design Studio San Francisco, CA, USA studio@kyuche.com ABSTRACT The Biometric Tendency Recognition and Classification System is a software application that continuously measures a participant’s physiological responses to a given image and runs a statistical classification algorithm on the measured data that then classifies the participant into one of four categories: passive, aggressive, loyal and subversive. The system is part of an interactive art project that explores issues of authority, privacy and security in relation to biometric technologies. In this paper, we demonstrate the development of this system that exemplifies the use of biometrics within the context of art. Categories and Subject Descriptors J.5 [Arts and Humanities]: fine arts. General Terms Algorithms, Performance, Design, Experimentation, Security, Human Factors. Keywords Biometrics, Biofeedback, Physiological Monitoring, Interactive Art, Machine Learning, K-Nearest Neighbor. 1. INTRODUCTION As biometric technologies become more prominent and accepted as part of everyday life, bio-data is becoming a legitimate source of identity detection, classification and assessment. What is an individuals relationship to this technology? How private is one's physiological data? How much are we willing to give up to intelligent machines in the name of safety or convenience? From another perspective, what happens when the human body and its emotional and cognitive responses become part of a sophisticated information system? In this paper, we present the Biometric Tendency Recognition and Classification System (henceforth referred to as BTRCS). This system is part of the technological foundation for Project Naos, an interactive art project that explores the use of physiological and biometric data in society. The approach taken with BTRCS is that of a hybrid art/science research project. In systems where there is a high degree of uncertainty and subjectivity, where specific goals are not well defined (or are constantly changing), we believe that new approaches to knowledge generation become necessary. In the case of data mining and machine learning techniques, which attempt to quantitatively ascertain qualitative and highly subjective characteristics, the opportunity for interpretation, manipulation and even distortion seem endless. Thus, opportunities for artistic “intervention” are high. The remainder of this paper is structured as follows. First, a general overview of Project Naos is provided and related works are discussed. We then present the system’s functionality, a description of the physiological parameters being measured, a discussion of the image database and a description of the classification algorithm used. A software implementation is also discussed which illustrates the use of the system in classifying participants based on their responses to the images displayed. We then explain the uses and limitations of the system and conclude with some possible future directions. 2. PROJECT NAOS 1 Project Naos is an interactive art project that explores possible future scenarios where the cognitive and emotional states of humans are monitored, profiled and classified based on their physiologial states. Building on the success of the Acclair project [1], of which it is an extension, Project Naos takes its inspiration from the various data mining and machine learning technologies that for example, attempt to determine categories of consumers, or to ascertain if one is a “security threat”. As shown in Figure 1, the installation consists of a “bio-pod” structure that participants enter and have their physiological responses to media measured in order to determine their social value and security status. A typical participant experience of the artwork is as follows: the participant approaches the Project Naos service station where an attendant greets him or her and explains how the process works. The participant is then directed into the bio-pod where he or she is connected to the physiological monitoring equipment. Once the participant is placed (un)comfortably inside the bio-pod, the 1 “Naos” is taken from the Pinyin spelling of the Mandarin word nao3, which means “brain”. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. DIMEA’08, September 10–12, 2008, Athens, Greece. Copyright 2008 ACM 978-1-60558-248-1/08/09…$5.00.