Publication View

Dynamic Nonlocal Language Modeling via Hierarchical Topic-Based Adaptation (1999)

Abstract
This paper presents a novel method of generating and applying hierarchical, dynamic topic-based language models. It proposes and evaluates new cluster generation, hierarchical smoothing and adaptive topic-probability estimation techniques. These combined models help capture long-distance lexical dependencies. Experiments on the Broadcast News corpus show significant improvement in perplexity (10.5% overall and 33.5% on target vocabulary).

Publication details
Download http://citeseerx.ist.psu.edu/viewdoc/summary?doi=?doi=10.1.1.38.8736
Source http://bigram.cs.jhu.edu/~rflorian/papers/acl99.ps
Contributors CiteSeerX
Repository CiteSeerX - Scientific Literature Digital Library and Search Engine (United States)
Type text
Language English
Relation 10.1.1.34.8794, 10.1.1.14.8678, 10.1.1.10.7323, 10.1.1.34.8552, 10.1.1.19.5240, 10.1.1.57.2164, 10.1.1.68.2505, 10.1.1.32.366, 10.1.1.24.9889, 10.1.1.6.4245, 10.1.1.72.9857, 10.1.1.127.9747, 10.1.1.71.2797, 10.1.1.72.9476, 10.1.1.87.1331, 10.1.1.113.204, 10.1.1.137.6742, 10.1.1.28.1910