Abstract not found

Confidence Estimation has been extensively used in Speech Recognition and now it is also being applied in Statistical Machine Translation. Its basic goal is to estimate a confidence measure for each word in a given hypothesis, in order to locate those words, if any, that are likely to be incorrectly recognised or translated. It can be seen as a two-class pattern recognition problem in which each hypothesized word is transformed into a feature vector and then classified as either correct or incorrect. This view provides a solid, well-know framework, within which accurate dichotomizers (two-class classifiers) can be derived. In this paper, we study the performance of certain pattern features along with a smoothed Naive Bayes dichotomizer. Good empirical results are reported on a translation task of technical manuals.

We present a detailed study of confidence estimation for machine translation. Various methods for determining whether MT output is correct are investigated, for both whole sentences and words. Since the notion of correctness is not intuitively clear in this context, different ways of defining it are proposed. We present results on data from the NIST 2003 Chinese-to-English MT evaluation. 1

entropy approaches

The goal of Utterance Verification is to estimate a confidence measure which helps detecting words in the hypothesized sentence that are likely to have been missrecognized. Word graphs have been extensively employed for directly estimating the confidence measure and for extracting important predictor features. In all the cases, a single word graph which is obtained through the recognition process. In this paper we propose the use of multiple word graphs to compute new features. The experimental study shows that these proposed features outperform those computed on a single word graph and other well-known predictor features. Moreover, the combination of the proposed features along with other kind of features provides improvements in the verification accuracy. 1.

In this work we implement a confidence estimation system based on a Naive Bayes classifier, by using the maximum entropy paradigm. The model takes information from various sources including a set of scores which have proved to be useful in confidence estimation tasks. Two different approaches are modeled. First a basic model which takes advantages of smoothing techniques used in a previous work, and second an optimized model, which is designed to hold a set of very few but essential characteristics of the model, without decrease in the performance. A considerably reduction in the number of parameters is obtained compared to the basic model. Both models are evaluated with two different corpora and compared to a model previously developed. Index Terms — confidence estimation, maximum entropy, confidence measures, speech recognition. 1.