Title :
Resolution of Pronominals in Tamil
Author_Institution :
MIT, Anna Univ., Chennai
Abstract :
The paper presents a system, which resolves pronominals in Tamil. In resolving the pronominals we adopt a probabilistic method. The analysis depends on the salience weight of the candidate noun phrases (NP) for the antecedent-hood of the pronominal from the list of possible candidate NPs. The salience weight of an NP is obtained from the salience factors, which are determined by the probability of an NP to be the antecedent on the basis of the grammatical features. The input text is pre-processed and partially parsed before sending it to the salience analyzer. The partial parsing is done using morphological markings and no sophisticated notions of the modern formal linguistic theories are used. The system considers candidate nouns that occur four sentences above the sentence in which the pronoun occurs. The system gives 86.32% precision and 80.9% recall
Keywords :
computational linguistics; grammars; natural language processing; probability; Tamil; formal linguistic theory; noun phrase; probabilistic method; Calculus; Computer applications; Dairy products; Machine learning; Machine learning algorithms; Natural languages; Pediatrics; Statistical analysis; Turning;
Conference_Titel :
Computing: Theory and Applications, 2007. ICCTA '07. International Conference on
Conference_Location :
Kolkata
Print_ISBN :
0-7695-2770-1
DOI :
10.1109/ICCTA.2007.112