Söz dizim ve anlam benzerliğine dayalı sözcük anlamı belirleme

M.S. Thesis Abstract FormName of the Student : Basak MutlumProgram of Study : Computer EngineeringThesis Title : Word Sense Disambiguation Based on SenseSimilarity and Syntactic ContextAbstractWord Sense Disambiguation (WSD) is the task of determining the meaning ofan ambiguous word within a given context. It is an open problem that has to besolved effectively in order to meet the needs of other natural languageprocessing tasks. Supervised and unsupervised algorithms have been triedthroughout the WSD research history. Up to now, supervised systems achievedthe best accuracies. However, these systems with the first sense heuristic havecome to a natural limit. In order to make improvement in WSD, benefits ofunsupervised systems should be examined.In this thesis, an unsupervised algorithm based on sense similarity andsyntactic context is presented. The algorithm relies on the intuition that twodifferent words are likely to have similar meanings if they occur in similar localcontexts. With the help of a principle-based broad coverage parser, a 100-million-word training corpus is parsed and local context features are extractedbased on some rules. Similarity values between the ambiguous word and thewords that occurred in a similar local context as the ambiguous word areevaluated. Based on a similarity maximization algorithm, polysemous words aredisambiguated. The performance of the algorithm is tested on SENSEVAL-2 andSENSEVAL-3 English all-words task data and an accuracy of 59% is obtained.Advisor: Deniz Yuret Date: 22.09.2005Director: Date: