Text classification aims at mapping documents into a set of predefined categories. Supervised machine learning models have shown great success in this area but they require a large number of labeled documents to reach adequate accuracy. This is particularly true when the number of target categories is in the tens or the hundreds. In this work, we explore an unsupervised approach to classify documents into categories simply described by a label. The proposed method is inspired by the way a human proceeds in this situation: It draws on textual similarity between the most relevant words in each document and a dictionary of keywords for each category reflecting its semantics and lexical field. The novelty of our method hinges on the enrichment of the category labels through a combination of human expertise and language models, both generic and domain specific. Our experiments on 5 standard corpora show that the proposed method increases F1-score over relying solely on human expertise and can also be on par with simple supervised approaches. It thus provides a practical alternative to situations where low cost text categorization is needed, as we illustrate with our application to operational risk incidents classification.

This paper investigates how to improve performance on information extraction tasks by constraining and sequencing CRF-based approaches.We consider two different relation extraction tasks, both from the medical literature: dependence relations and probability statements.We explore whether adding constraints can lead to an improvement over standard CRF decoding.Results on our relation extraction tasks are promising, showing significant increases in performance from both (i) adding constraints to post-process the output of a baseline CRF, which captures "domain knowledge", and (ii) further allowing flexibility in the application of those constraints by leveraging a binary classifier as a pre-processing step.

Building Bayesian belief networks in the absence of data involves the challenging task of eliciting conditional probabilities from experts to parameterize the model. In this paper, we develop an analytical method for determining the optimal order for eliciting these probabilities. Our method uses prior distributions on network parameters and a novel expected proximity criteria, to propose an order that maximizes information gain per unit elicitation time. We present analytical results when priors are uniform Dirichlet; for other priors, we find through experiments that the optimal order is strongly affected by which variables are of primary interest to the analyst. Our results should prove useful to researchers and practitioners involved in belief network model building and elicitation.