|
|
|
                                          
|
Return to Posters
A machine learning methodology for the disambiguation of acronym senses is presented, which starts from an acronym sense dictionary. Training data is automatically extracted from downloaded documents identified from the results of search engine queries. Leave-one-out cross-validation on 9,963 documents with 47 acronym forms achieves accuracy 92.58% and Fß=1=91.52%.
@inproceedings{1009133, author = {Manuel Zahariev}, title = {Automatic sense disambiguation for acronyms}, booktitle = {SIGIR '04: Proceedings of the 27th annual international conference on Research and development in information retrieval}, year = {2004}, isbn = {1-58113-881-4}, pages = {586--587}, location = {Sheffield, United Kingdom}, doi = {http://doi.acm.org/10.1145/1008992.1009133}, publisher = {ACM Press}, }  ©2005 Association for Computing Machinery |
