The quality of a decision tree is usually evaluated through its complexity and its generalization accuracy. Tree-simpliÞcation procedures aim at optimizing these two performance criteria. Among them, data reduction techniques differ from pruning by their simpliÞcation strategy. Actually, while pruning algorithms directly control tree size to combat the overÞtting problem, data reduction techniq...

T concept of dominance among nodes of a branch-and-bound tree, although known for a long time, is typically not exploited by general-purpose mixed-integer linear programming (MILP) codes. The starting point of our work was the general-purpose dominance procedure proposed in the 1980s by Fischetti and Toth, where the dominance test at a given node of the branch-and-bound tree consists of the (po...

Non-negative matrix based language models have been recently introduced [1] as a computationally efficient alternative to other feature-based models such as maximum-entropy models. We present a new entropy based pruning algorithm for this class of language models, which is fast and scalable. We present perplexity and word error rate results and compare these against regular n-gram pruning. We a...

We study the following question: when can a mined pattern, which may be an association, a correlation, ratio rule, or any other, be regarded as interesting? Previous approaches to answering this question have been largely numeric. Speciically, we show that the presence of some rules may make others redundant, and therefore uninteresting. We articulate these principles and formalize them in the ...

It has been asserted that, using traditional pruning methods, growing decision trees with increasingly larger amounts of training data will result in larger tree sizes even when accuracy does not increase. With regard to error-based pruning, the experimental data used to illustrate this assertion have apparently been obtained using the default setting for pruning strength; in particular, using ...

An efficient decoding algorithm for segmental HMMs (SHMMs) is proposed with multi-stage pruning. The generation by SHMMs of a feature trajectory for each state expands the search space and the computational cost of decoding. It is reduced in three ways: pre-cost partitioning, start-node (SN) beam pruning, and conventional endnode (EN) beam pruning. Experiments show that partitioning cuts comput...