Husrev Tolga Ilhan

Learn More
This paper describes improved HMM-based word level alignment models for statistical machine translation. We present a method for using part of speech tag information to improve alignment accuracy, and an approach to modeling fertility and correspondence to the empty word in an HMM alignment model. We present accuracy results from evaluating Viterbi(More)
This paper discusses ensembles of simple but heterogeneous classifiers for word-sense disambiguation, examining the Stanford-CS224N system entered in the SENSEVAL-2 English lexical sample task. First-order classifiers are combined by a second-order classifier, which variously uses majority voting, weighted voting, or a maximum entropy model. While(More)
This paper describes improved HMM-based word level alignment models for statistical machine translation. We present a method for using part of speech tag information to improve alignment accuracy, and an approach to modeling fertility and correspondence to the empty word in an HMM alignment model. We present accuracy results from evaluating Viterbi(More)
We present an approach to the common problem in data analysis namely to detect interesting phenomena without a priori knowledge. In an earlier work [1,2] we put forth method of decomposition in one dimension based on a notion we called degree of symmetry. In this work we present a decomposition model in two dimensions based as well on degree of symmetry. We(More)
  • 1