The efficacy of a machine learning technique is domain dependent. Some machine learning techniques work very well for certain domains but are ill-suited for other domains. One area that is of real-world concern is the flexibility with which machine learning techniques can adapt to dynamic domains. Currently, there are no known reports of any system that can learn dynamic domains, short of starting over (i.e., re-running the program). Starting over is neither time nor cost efficient for real-world production environments. This dissertation studied a method, referred to as Experience Based Learning (EBL), that attempts to deal with conditions related to learning dynamic domains. EBL is an extension of Instance Based Learning methods. The hypothesis of the study related to this research was that the EBL method would automatically adjust to domain changes and still provide classification accuracy similar to methods that require starting over. To test this hypothesis, twelve widely studied machine learning datasets were used. A dynamic domain was simulated by presenting these datasets in an uninterrupted cycle of train, test, and retrain. The order of the twelve datasets and the order of records within each dataset were randomized to control for order biases in each of ten runs. As a result, these methods provided datasets that represent extreme levels of domain change. Using the above datasets, EBL's mean classification accuracies for each dataset were compared to the published static domain results of other machine learning systems. The results indicated that the EBL's system performance was not statistically different (p>0.30) from the other machine learning methods. These results indicate that the EBL system is able to adjust to an extreme level of domain change and yet produce satisfactory results. This finding supports the use of the EBL method in real-world environments that incur rapid changes to both variables and values.

No description available.

No description available.

Our motivation for this work is to develop an autonomous robot system that is able to perform autism intervention therapy. Autism spectrum disorder (ASD) is a common type of neurodevelopmental disorder that affects millions of people in the United States alone. The best way of treating ASD and help people with ASD learn new skills is through applied behavior analysis (ABA, i.e. autism intervention therapy). Because of the fact that people with ASD feel less stressful in a predictable and simple environment compared to interacting with other people and autism intervention therapy provided by professional therapists are generally expensive and inaccessible, it would be beneficial to build robots that can perform intervention therapy with children without a therapist/instructor present. In this research, we focus on the task of detecting engagement/disengagement levels of a child in a therapy session as a first step in designing a therapy robot. In this work, we mainly utilize an RGB-D camera, namely the Microsoft Kinect 2.0, to extract kinematic joint data from the therapy session. We also set up a child study with the Kid’s Creek therapy center to recruit children with ASD and record their interactions with a therapist while working on a touch-screen based game on a tablet. After carefully selecting features derived from skeletons’ movements and poses, we showed that our system can produce an accuracy of 97% when detecting engagements and disengagements using cross-validation assessment.

No description available.

No description available.

No description available.

No description available.

No description available.

This dissertation explores how decisions about the forecasting process can affect the evaluation of forecasting performance, in general and in the domain of residential electricity demand estimation. Decisions of interest include those around data sourcing, sampling, clustering, temporal magnification, algorithm selection, testing approach, evaluation metrics, and others. Models of the forecasting process and analysis methods are formulated in terms of a three-tier decision taxonomy, by which decision effects are exposed through systematic enumeration of the techniques resulting from those decisions. A computation platform based on the models is implemented to compute and visualize the effects. The methods and computation platform are first demonstrated by applying them to 3,003 benchmark datasets to investigate various decisions, including those that could impact the relationship between data entropy and forecastability. Then, they are used to study over 10,624 week-ahead and day-ahead residential electricity demand forecasting techniques, utilizing fine-resolution electricity usage data collected over 18 months on groups of 782 and 223 households by real smart electric grids in Ireland and Australia, respectively. The main finding from this research is that forecasting performance is highly sensitive to the interaction effects of many decisions. Sampling is found to be an especially effective data strategy, clustering not so, temporal magnification mixed. Other relationships between certain decisions and performance are surfaced, too. While these findings are empirical and specific to one practically scoped investigation, they are potentially generalizable, with implications for residential electricity demand estimation, smart electric grid design, and electricity policy.