This paper describes a variety of nonparametric Bayesian models of word segmentation based on Adaptor Grammars that model different aspects of the input and incorporate different kinds of prior knowledge, and applies them to the Bantu language Sesotho. While we find overall word segmentation accuracies lower than these models achieve on English, we also find some interesting differences in whic...

In this paper, we describe a new unsupervised sentence boundary detection system and present a comparative study evaluating its performance against different systems found in the literature that have been used to perform the task of automatic text segmentation into sentences for English and Portuguese documents. The results achieved by this new approach were as good as those of the previous sys...

Most work on language acquisition treats word segmentation—the identification of linguistic segments from continuous speech— and word learning—the mapping of those segments to meanings—as separate problems. These two abilities develop in parallel, however, raising the question of whether they might interact. To explore the question, we present a new Bayesian segmentation model that incorporates...

0

This paper proposes a new method of Chinese word segmentation based on proportional analogy and majority voting. First, we introduce an analogy-based method for solving the word segmentation problem. Second, we show how to use majority voting to make the decision on where to segment. The preliminary results show that this approach compares well with other segmenters reported in previous studies...

This paper explores the segmentation of tutorial dialogue into cohesive topics. A latent semantic space was created using conversations from human to human tutoring transcripts, allowing cohesion between utterances to be measured using vector similarity. Previous cohesionbased segmentation methods that focus on expository monologue are reapplied to these dialogues to create benchmarks for perfo...

Evidence that I presented at the last meeting in Yokohama indicated that English-learning infants first show some capacity for segmenting words from fluent speech at about 7.5 months of age. Further studies that we have conducted suggest that English-learning infants initially rely on a prosodically based strategy which may cause them to mis-segment words beginning with weak syllables. However,...