Machine learning for text classification is the cornerstone of document categorization, news filtering, document routing, and personalization. In text domains, effective feature selection is essential to make the learning task efficient and more accurate. This paper presents an empirical comparison of twelve feature selection methods (e.g. Information Gain) evaluated on a benchmark of 229 text classification problem instances that were gathered from Reuters, TREC, OHSUMED, etc. The results are analyzed from multiple goal perspectives—accuracy, F-measure, precision, and recall—since each is appropriate in different situations. The results reveal that a new feature selection metric we call ‘Bi-Normal Separation ’ (BNS), outperformed the others by a substantial margin in most situations. This margin widened in tasks with high class skew, which is rampant in text classification problems and is particularly challenging for induction algorithms. A new evaluation methodology is offered that focuses on the needs of the data mining practitioner faced with a single dataset who seeks to choose one (or a pair of) metrics that are most likely to yield the best performance. From this perspective, BNS was the top single choice for all goals except precision, for which Information Gain yielded the best result most often. This analysis also revealed, for example, that Information Gain and Chi-Squared have correlated failures, and so they work poorly together. When choosing optimal pairs of metrics for each of the four performance goals, BNS is consistently a member of the pair—e.g., for greatest recall, the pair BNS + F1-measure yielded the best performance on the greatest number of tasks by a considerable margin.

Models of document indexing and document retrieval have been extensively studied. The integration of these two classes of models has been the goal of several researchers but it is a very difficult problem. We argue that much of the reason for this is the lack of an adequate indexing model

We present an extensive empirical comparison of several smoothing techniques in the domain of language modeling, including those described by Jelinek and Mercer (1980), Katz (1987), and Church and Gale (1991). We investigate for the first time how factors such as training data size, corpus (e

Miller [12] for a general discussion and bibliography). Events ordered in time are particularly amenable to informational analysis; thus language sequences are being extensively studied, and other sequences, such as those of music, plainly invite research. In this paper I shall indicate some of the ways

using finitely many real-valued clock variables. Automated analysis of timed automata relies on the construction of a finite quotient of the infinite space of clock valuations. Over the years, the formalism has been extensively studied leading to many results establishing connections to circuits

the influence of background correlations on recognition performance via extensive tests on the PASCAL database, for which ground-truth object localization information is available. Our experiments demonstrate that image representations based on distributions of local features are surprisingly effective

We introduce the spi calculus, an extension of the pi calculus designed for the description and analysis of cryptographic protocols. We show how to use the spi calculus, particularly for studying authentication protocols. The pi calculus (without extension) suffices for some abstract protocols

layer. We present our observations based on an extensive series of experiments, draw conclusions from a wide range of scenarios, and are more cautious and less optimistic as compared to previous studies.

, including several which explicitly learn the correspondence between regions and words. We study multi-modal and correspondence extensions to Hofmann’s hierarchical clustering/aspect model, a translation model adapted from statistical machine translation (Brown et al.), and a multi-modal extension to mixture

Reigning measures of psychological well-being have little theoretical grounding, despite an extensive literature on the contours of positive functioning. Aspects of well-being derived from this literature (i.e., self-acceptance, positive relations with others, autonomy, environmental mastery