This paper describes a framework for defining domain specific Feature Functions in a user friendly form to be used in a Maximum Entropy Markov Model (MEMM) for the Named Entity Re...
Existing Recurrent Neural Networks (RNNs) are limited in their ability to model dynamical systems with nonlinearities and hidden internal states. Here we use our general framework...
Speech-based applications commonly come with web-based or printed manuals. Alternatively, the dialogue can be designed so that users should be able to start using the application ...
Jaakko Hakulinen, Markku Turunen, Esa-Pekka Salone...
We demonstrate that regularization can improve feedback in a language modeling framework. Categories and Subject Descriptors: H.3.3 Information Search and Retrieval: Relevance Fee...
Higman showed that if A is any language then SUBSEQ(A) is regular, where SUBSEQ(A) is the language of all subsequences of strings in A. We consider the following inductive inferenc...