Summary of the paper
| Title | Handling of Missing Values in Lexical Acquisition |
|---|---|
| Authors | Núria Bel |
| Abstract | In this work we propose a strategy to reduce the impact of the sparse dataproblem in the tasks of lexical information acquisition based on theobservation of linguistic cues. We propose a way to handle the uncertaintycreated by missing values, that is, when a zero value could mean either thatthe cue has not been observed because the word in question does not belong tothe class, i.e. negative evidence, or that the word in question has just notbeen observed in the context sought by chance, i.e. lack of evidence. Thisuncertainty creates problems to the learner, because zero values forincompatible labelled examples make the cue lose its predictive capacity andeven though some samples display the sought context, it is not taken intoaccount. In this paper we present the results of our experiments to try to reduce thisuncertainty by, as other authors do (Joanis et al. 2007, for instance),substituting zero values for pre-processed estimates. Here we present a firstround of experiments that have been the basis for the estimates of linguisticinformation motivated by lexical classes. We obtained experimental results thatshow a clear benefit of the proposed approach. |
| Language | Statistical and machine learning methods |
| Topics | Lexicon, lexical database, Acquisition, Statistical and machine learning methods |
Full paper  |
Handling of Missing Values in Lexical Acquisition |
| Bibtex | @InProceedings{BEL10.45, author = {Núria Bel}, title = {Handling of Missing Values in Lexical Acquisition}, booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)}, year = {2010}, month = {may}, date = {19-21}, address = {Valletta, Malta}, editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias}, publisher = {European Language Resources Association (ELRA)}, isbn = {2-9517408-6-7}, language = {english} } |