Protein secondary structure prediction plays a pivotal role in predicting protein folding in three-dimensions. Its task is to assign each residue one of the three secondary structure classes helix, strand, or random coil. This is an instance of the problem of sequential supervised learning in machine learning. This thesis describes...
Knowledge workers are struggling in the information flood. There is a growing interest in intelligent desktop environments that help knowledge workers organize their daily life. Intelligent desktop environments allow the desktop user to define a set of “activities” that characterize the user’s desktop work. These environments then attempt to identify...
Simulation has provided much design inspiration for the field of robotics. While many astute ideas can be computationally formulated, there are some good structures in animals that people call “biologically-inspired”. Many robots are designed based on natural creatures such as crabs, spiders, etc. Before we can take full advantage of...
A new system called sequential/parallel matrix grammars
for two-dimensional pattern processing is introduced and studied.
Miscellaneous language operations such as union, catenation (row
and column), Kleene's closure (row and column) and substitutions
are investigated. The equivalence of sequential/parallel matrix
languages and finite-turn repetitive checking automata is established.
Hierarchies for both...
This paper examines how six online multiclass text classification algorithms perform in the domain of email tagging within the TaskTracer system. TaskTracer is a project-oriented user interface for the desktop knowledge worker. TaskTracer attempts to tag all documents, web pages, and email messages with the projects to which they are...
This dissertation delves into understanding, characterizing, and addressing dataset shift in deep learning, a pervasive issue for deployed machine learning systems. Integral aspects of the problem are examined: We start with the use of counterfactual explanations in order to characterize the behavior of deep reinforcement learning agents in visual input...
This paper reviews McCabe's cyclomatic complexity and Halstead's laws; it discusses studies in current literature relating the metrics to software. The studies are reproduced using data obtained from a large software project developed in a major electronics firm. Problems that occur when deriving the metrics are discussed; the result of...
AM is a computer program written by Doug Lenat that discovers elementary
mathematics starting from some initial knowledge of set theory. The success of this
program is not clearly understood.
This work is an attempt to explore the search space of AM in order to understand the success and eventual...
This dissertation explores the idea of applying machine learning technologies to help computer users find information and better organize electronic resources, by presenting the research work conducted in the following three applications: FolderPredictor, Stacking Recommendation Engines, and Integrating Learning and Reasoning.
FolderPredictor is an intelligent desktop software tool that helps...