Abstract. In this work, algorithms are developed and evaluated to detect physical activities from data acquired using five small biaxial accelerometers worn simultaneously on different parts of the body. Acceleration data was collected from 20 subjects without researcher supervision or observation. Subjects were asked to perform a sequence of everyday tasks but not told specifically where or how to do them. Mean, energy, frequency-domain entropy, and correlation of acceleration data was calculated and several classifiers using these features were tested. Decision tree classifiers showed the best performance recognizing everyday activities with an overall accuracy rate of 84%. The results show that although some activities are recognized well with subject-independent training data, others appear to require subject-specific training data. The results suggest that multiple accelerometers aid in recognition because conjunctions in acceleration feature values can effectively discriminate many activities. With just two biaxial accelerometers -thigh and wrist -the recognition performance dropped only slightly. This is the first work to investigate performance of recognition algorithms with multiple, wire-free accelerometers on 20 activities using datasets annotated by the subjects themselves.

Abstract. Three tools for acquiring data about people, their behavior, and their use of technology in natural settings are described: (1) a context-aware experience sampling tool, (2) a ubiquitous sensing system that detects environmental changes, and (3) an image-based experience sampling system. We discuss how these tools provide researchers with a flexible toolkit for collecting data on activity in homes and workplaces, particularly when used in combination. We outline several ongoing studies to illustrate the versatility of these tools. Two of the tools are currently available to other researchers to use.

A new software tool for user-interface development and assessment of ubiquitous computing applications is available for CHI researchers. The software permits researchers to use common PDA mobile computing devices for experience sampling studies. The basic tool offers options not currently available in any other open-source sampling package. However, the tool also has one a completely new type of functionality: context-aware experience sampling. This feature permits researchers to acquire feedback from users only in particular situations that are detected by sensors connected to a mobile computing device. KeywordsContext-aware, experience sampling, reflection, eliciting preferences, PDA, ubiquitous and mobile computing. THE PROBLEMUser needs are typically elicited via personal or focus group interviews, site visits, and photographic and video analysis. Often, however, users know more than they say in a single or even several interviews [1]. As user interface design moves off the desktop and into the real world, two new challenges for designers emerge: (1) developing realistic task specifications that respond to the complexity of fastchanging, real world activities, and (2) evaluating new technologies in realistic contexts. Desktop computing applications can be designed and evaluated using controlled, laboratory observation because most user interface design has nothing to do with physical space [2]. Developers of ubiquitous and mobile computing applications for the home and workplace, however, currently lack a powerful and economical assessment toolset that accounts for user activity in a broader context. The behavior of the people and their response to technology is critically dependent upon the environment and context in which information is presented or requested.The most popular assessment instruments in use today for studying the activities of people in natural settings are self report recall surveys, time diaries, direct field observation, and experience sampling. Self-report recall surveys suffer from recall and selective reporting biases -users can often not remember what they did. Time diaries, where users write down what they do during the day, are burdensome for the user. Although direct field observation can provide helpful qualitative and quantitative measures, it is costly, timeconsuming, and disruptive and therefore not practical for many design tasks. The experience sampling method (ESM) has been used primarily for time-use analysis [3] and only recently for interface design [4,5]. Subjects carry a beeper device that "samples" for information on some predetermined schedule. When the device beeps, subjects answer questions of interest to the researchers. With a sufficient number of subjects and samples, a statistical model of behavior can be generated. The ESM is less susceptible to subject recall errors than other self-report feedback elicitation methods [3], but it high sampling rates interrupt activities of interest and irritate subjects. Image-based experience sampling alleviates these ...

Ubiquitous, context-aware computer systems may ultimately enable computer applications that naturally and usefully respond to a user's everyday activity. Although new algorithms that can automatically detect context from wearable and environmental sensor systems show promise, many of the most flexible and robust systems use probabilistic detection algorithms that require extensive libraries of training data with labeled examples. In this paper, we describe the need for such training data and some challenges we have identified when trying to collect it while testing three contextdetection systems for ubiquitous computing and mobile applications.

Many algorithms have been used to cluster genes measured by microarray across a time series. Instead of clustering, our goal was to compare all pairs of genes to determine whether there was evidence of a phase shift between them. We describe a technique where gene expression is treated as a discrete time-invariant signal, allowing the use of digital signal-processing tools, including power spectral density, coherence, and transfer gain and phase shift. We used these on a public RNA expression set of 2467 genes measured every 7 min for 119 min and found 18 putative associations. Two of these were known in the biomedical literature and may have been missed using correlation coefficients. Digital signal processing tools can be embedded and enhance existing clustering algorithms.