Showing papers by "Thomas G. Dietterich published in 1992"
01 Jan 1992
TL;DR: Experimental studies show that the earning performance of ID3 is greatly improved when these algorithms are used to process the training data by eliminating the irrelevant features from ID3''s consideration.
Abstract: This paper describes different methods for exact and approximate implementation of the MIN-FEATURES bias, which prefers consistent hypotheses definable over as few features as possible. This bias is useful for learning domains where many irrelevant features are present in the training data. We first introduce FOCUS-2, a new algorithm that exactly implements the MIN-FEATURES bias. This algorithm is empirically shown to be substantially faster than the FOCUS algorithm previously given in [Almuallim and Dietterich 91]. We then introduce the Mutual-Information-Greedy, Simple-Greedy and Weighted-Greedy Algorithms, which apply efficient heuristics for approximating the MIN-Features bias. These algorithms employ greedy heuristics that trade optimality for computational efficiency. Experimental studies show that the ;earning performance of ID3 is greatly improved when these algorithms are used to process the training data by eliminating the irrelevant features from ID3''s consideration. In particular, the Weighted-Greedy algorithm provides an excellent and efficient approximation of the MIN-Features bias.
129 citations
TL;DR: A data representation is used to record the history of the design as a sequence of design decisions and the resulting data base records the final specifications, the alternatives which were considered during the design process, and the designers' rationale for choosing the final design parameters.
Abstract: Collaborative design projects place additional burdens on design documentation practices. The literature on group design has repeatedly documented the existence of problems in design decision-making due to the unavailability of design information. This paper describes a data representation developed for collaborative mechanical design information. The data representation is used to record the history of the design as a sequence of design decisions. The resulting data base records the final specifications, the alternatives which were considered during the design process, and the designers' rationale for choosing the final design parameters. It is currently implemented in a computerized data base system under development at Oregon State University (OSU).
31 citations