Improving text classification for oral history archives with temporal domain knowledge Export

@inproceedings{citeulike:1510685, 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.}, address = {New York, NY, USA}, author = {Olsson, Scott J. and Oard, Douglas W.}, booktitle = {SIRIR '07: Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval}, citeulike-article-id = {1510685}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1277741.1277848}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1277741.1277848}, doi = {10.1145/1277741.1277848}, isbn = {9781595935977}, keywords = {classification--automatic, classification--text--automated, digital-archives, machine\_learning--supervised}, pages = {623--630}, posted-at = {2007-07-29 00:56:35}, priority = {4}, publisher = {ACM Press}, title = {Improving text classification for oral history archives with temporal domain knowledge}, url = {http://dx.doi.org/10.1145/1277741.1277848}, year = {2007} }

TY - CONF ID - citeulike:1510685 L3 - citeulike-article-id:1510685 N2 - 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. SN - 9781595935977 AD - New York, NY, USA EP - 630 TI - Improving text classification for oral history archives with temporal domain knowledge PB - ACM Press T2 - SIRIR '07: Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval SP - 623 KW - classification--automatic KW - classification--text--automated KW - digital-archives KW - machine_learning--supervised AU - Olsson, Scott AU - Oard, Douglas PY - 2007/// UR - http://dx.doi.org/10.1145/1277741.1277848 ER -