Improving text classification for oral history archives with temporal domain knowledge
Published 23 July 2007
J. Scott Olsson, Douglas W. Oard
Citations11
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
Two new techniques for increasing the accuracy oftopic label assignment to conversational speech from oral history interviews using supervised machine learning in conjunction with automatic speech recognition are described.
Abstract
This paper describes two new techniques for increasing the accuracy oftopic label assignment to conversational speech from oral history interviews using supervised machine learning in conjunction with automatic speech recognition. The first, time-shifted classification, leverages local sequence information from the order in which the story is told. The second, temporal label weighting, takes the complementary perspective by using the position within an interview to bias label assignment probabilities. These methods, when used in combination, yield between 6% and 15% relative improvements in classification accuracy using a clipped R-precision measure that models the utility of label sets as segment summaries in interactive speech retrieval applications.
Keywords
Computer Science
Statisctics and computing/Statistics and computingModern Applied Statistics with S
24,337 Citations2002W. N. Venables, B. D. Ripley
A guide to using S environments to perform statistical analyses providing both an introduction to the use of S and a course in modern statistical methods.
A Comparative Study on Feature Selection in Text Categorization
4,766 Citations1997Yiming Yang, Jan Pedersen
DF thresholding, the simplest method with the lowest cost in computation, can be reliably used instead of IG or CHI when the computation of these measures are too expensive, and strong correlations between the DF, IG and CHI values of a term are found.
Neural ComputationApproximate Statistical Tests for Comparing Supervised Classification Learning Algorithms
3,654 Citations1998Thomas G. Dietterich
This article reviews five approximate statistical tests for determining whether one learning algorithm outperforms another on a particular learning task and measures the power (ability to detect algorithm differences when they do exist) of these tests.
Text REtrieval ConferenceOkapi at TREC
2,229 Citations1994Stephen Robertson, Steve Walker +3 more
Much of the work involved investigating plausible methods of applying Okapi-style weighting to phrases, and expansion using terms from the top documents retrieved by a pilot search on topic terms was used.
Feature generation for text categorization using world knowledge
232 Citations2005Evgeniy Gabrilovich, Shaul Markovitch
Improved machine learning algorithms for text categorization with generated features based on domain-specific and common-sense knowledge are enhanced, addressing the two main problems of natural language processing--synonymy and polysemy.
IEEE Transactions on Speech and Audio ProcessingAutomatic Recognition of Spontaneous Speech for Access to Multilingual Oral History Archives
133 Citations2004Bill Byrne, David Doermann +10 more
Results are presented from experiments with speech recognition, topic segmentation, topic categorization, and named entity detection using a large collection of recorded oral histories to evaluate the degree to which automatic speech recognition (ASR)-based segmentation and categorization techniques can be adapted to approximate decisions made by human annotators.
Incorporating prior knowledge with weighted margin support vector machines
129 Citations2004Wu Xiaoyun, Rohini K. Srihari
A simple generalization of SVM: Weighted Margin SVM (WMSVMs) that permits the incorporation of prior knowledge and it is shown that Sequential Minimal Optimization can be used in training WMSVM.
Incorporating Prior Knowledge into Boosting
116 Citations2002Robert E. Schapire, Marie Rochery +2 more
Bootstrapping for text learning tasks
89 Citations1999Ellen Riloff
This paper presents bootstrapping as an alternative approach to learning from large sets of labeled data, using a small amount of seed information and a large collection of easily-obtained unlabeled data.
Tackling concept drift by temporal inductive transfer
88 Citations2006George Forman
A new learning model is developed that transfers induced knowledge through time to benefit future classifier learning tasks and avoids two main problems with existing work in inductive transfer: scalability and the risk of negative transfer.
Communications of the ACMIntegrated technologies for indexing spoken language
59 Citations2000Francis Kubala, Sean Colbath +3 more
After more than a decade of steady advances in speech recognition, speaker identification, and language understanding, it is possible to begin building usable automatic content-based indexing tools for spoken language by integrating these emerging technologies.
Constructing informative prior distributions from domain knowledge in text classification
54 Citations2006Aynur A. Dayanik, David Lewis +3 more
This work proposes combining domain knowledge with training examples in a Bayesian framework, and shows that this approach can rescue what would otherwise be disastrously bad training situations, producing much more effective classifiers.
Cross-language text classification
50 Citations2005J. Scott Olsson, Douglas W. Oard +1 more
The goal in cross-language text classification (CLTC) is to use English training data to classify Czech documents (although the concepts presented here are applicable to any language pair), and the authors are unaware of any previous work in this area.
Lecture notes in computer scienceOverview of the CLEF-2006 Cross-Language Speech Retrieval Track
47 Citations2007Douglas W. Oard, Jian‐qiang Wang +6 more
Lecture notes in computer scienceSearch of Spoken Documents Retrieves Well Recognized Transcripts
11 Citations2007Mark Sanderson, Xiao Mang Shou
Analysis showing a relationship between word error rate and query length and results show that transcripts that match queries well tend to be recognized more accurately than transcripts thatMatch a query less well.