Automatic topic detection from news stories.

Hui Kin.Thesis (M.Phil.)--Chinese University of Hong Kong, 2001.Includes bibliographical references (leaves 115-120).Abstracts in English and Chinese.Chapter 1 --- Introduction --- p.1Chapter 1.1 --- Topic Detection Problem --- p.2Chapter 1.1.1 --- What is a Topic? --- p.2Chapter 1.1.2 --- Topic Detection --- p.3Chapter 1.2 --- Our Contributions --- p.5Chapter 1.2.1 --- Thesis Organization --- p.6Chapter 2 --- Literature Review --- p.7Chapter 2.1 --- Dragon Systems --- p.7Chapter 2.2 --- University of Massachusetts (UMass) --- p.9Chapter 2.3 --- Carnegie Mellon University (CMU) --- p.10Chapter 2.4 --- BBN Technologies --- p.11Chapter 2.5 --- IBM T. J. Watson Research Center --- p.12Chapter 2.6 --- National Taiwan University (NTU) --- p.13Chapter 2.7 --- Drawbacks of Existing Approaches --- p.14Chapter 3 --- System Overview --- p.16Chapter 3.1 --- News Sources --- p.17Chapter 3.2 --- Story Preprocessing --- p.21Chapter 3.3 --- Named Entity Extraction --- p.22Chapter 3.4 --- Gross Translation --- p.22Chapter 3.5 --- Unsupervised Learning Module --- p.24Chapter 4 --- Term Extraction and Story Representation --- p.27Chapter 4.1 --- IBM Intelligent Miner For Text --- p.28Chapter 4.2 --- Transformation-based Error-driven Learning --- p.31Chapter 4.2.1 --- Learning Stage --- p.32Chapter 4.2.2 --- Design of New Tags --- p.33Chapter 4.2.3 --- Lexical Rules Learning --- p.35Chapter 4.2.4 --- Contextual Rules Learning --- p.39Chapter 4.3 --- Extracting Named Entities Using Learned Rules --- p.42Chapter 4.4 --- Story Representation --- p.46Chapter 4.4.1 --- Basic Representation --- p.46Chapter 4.4.2 --- Enhanced Representation --- p.47Chapter 5 --- Gross Translation --- p.52Chapter 5.1 --- Basic Translation --- p.52Chapter 5.2 --- Enhanced Translation --- p.60Chapter 5.2.1 --- Parallel Corpus Alignment Approach --- p.60Chapter 5.2.2 --- Enhanced Translation Approach --- p.62Chapter 6 --- Unsupervised Learning Module --- p.68Chapter 6.1 --- Overview of the Discovery Algorithm --- p.68Chapter 6.2 --- Topic Representation --- p.70Chapter 6.3 --- Similarity Calculation --- p.72Chapter 6.3.1 --- Similarity Score Calculation --- p.72Chapter 6.3.2 --- Time Adjustment Scheme --- p.74Chapter 6.3.3 --- Language Normalization Scheme --- p.75Chapter 6.4 --- Related Elements Combination --- p.78Chapter 7 --- Experimental Results and Analysis --- p.84Chapter 7.1 --- TDT corpora --- p.84Chapter 7.2 --- Evaluation Methodology --- p.85Chapter 7.3 --- Experimental Results on Various Parameter Settings --- p.88Chapter 7.4 --- Experiments Results on Various Named Entity Extraction Ap- proaches --- p.89Chapter 7.5 --- Experiments Results on Various Story Representation Approaches --- p.100Chapter 7.6 --- Experiments Results on Various Translation Approaches --- p.104Chapter 7.7 --- Experiments Results on the Effect of Language Normalization Scheme on Detection Approaches --- p.106Chapter 7.8 --- TDT2000 Topic Detection Result --- p.110Chapter 8 --- Conclusions and Future Works --- p.112Chapter 8.1 --- Conclusions --- p.112Chapter 8.2 --- Future Work --- p.114Bibliography --- p.115Chapter A --- List of Topics annotated for TDT2 Corpus --- p.121Chapter B --- Significant Test Results --- p.12