This paper presents an unsupervised method that uses limited amount of labeled data and a large pool of unlabeled data to improve natural language call routing performance. The method uses multiple classifiers to select a subset of the unlabeled data to augment limited labeled data. We evaluated four widely used text classification algorithms; Naive Bayes Classification (NBC), Support Vector ma...

We propose a novel reordering model for phrase-based statistical machine translation (SMT) that uses a maximum entropy (MaxEnt) model to predicate reorderings of neighbor blocks (phrase pairs). The model provides content-dependent, hierarchical phrasal reordering with generalization based on features automatically learned from a real-world bitext. We present an algorithm to extract all reorderi...

We derive a dual-primal recursive algorithm based on the Fenchel duality framework, extending Dykstra’s successive projections and Csiszar’s I-projections schemes, to handle Tsallis MaxEnt. The Tsallis entropy Sq(p) is a one-parameter extension of Shannon’s entropyH(p) in the sense that Sq→1(p) = H(p). The solution of the Tsallis MaxEnt falls under a q-deformed Gibbs distribution which is a pow...

Various supervised inference methods can be analyzed as convex duals of the generalized maximum entropy (MaxEnt) framework. Generalized MaxEnt aims to find a distribution that maximizes an entropy function while respecting prior information represented as potential functions in miscellaneous forms of constraints and/or penalties. We extend this framework to semi-supervised learning by incorpora...

A single and self-contained method for determining the crystallite-size distribution and shape from experimental x-ray line profile data is presented. We have shown that the crystallite-size distribution can be determined without assuming a functional form for the size distribution, determining instead the size distribution with the least assumptions by applying the Bayesian/MaxEnt method. The ...

This paper focuses on verb sense disambiguation cast as inferring the VerbNet class to which a verb belongs. To train three different supervised learning models –Maximum Entropy (MaxEnt), Naive Bayes and Decision Tree– we used lexical, co-occurrence and typed-dependency features. For each model, we built three classifiers: one single classifier for all verbs, one single classifier for polysemou...

The paper discusses two policies for recognizing NEs with complex structures by maximum entropy models. One policy is to develop cascaded MaxEnt models at different levels. The other is to design more detailed tags with human knowledge in order to represent complex structures. The experiments on Chinese organization names recognition indicate that layered structures result in more accurate mode...