Afaan Oromo Parser using Hybrid Approach

Abstract

Nowadays, Natural Language Processing (NLP) concerns with the interaction between computers and human natural languages. The most difficult task in NLP is to learn natural languages for the computer. Parsing is one of the very important tasks in natural language processing. It is the task of analyzing the structural relationship between the words in a sentence. For a free word order language like Afaan Oromo, parser suits the best to extract the relation between the words in the sentences. Development of hybrid sentence parser for Afaan Oromo will avoid the large amount of time wasted to manually process sentences in the language to show its syntactic structure. The parser is also useful for semantic parsing which extracting meaning from a sentence and checking the well-formed-ness of a sentence, which is useful in a number of applications such as language teaching. Corpus used in this study as training and test set are manually parsed by researchers with linguistic advisor. Manually parsed sentences are given to machine for machine learning. In this thesis, Weka tool is used for machine learning technique. The algorithm used for machine learning is support vector machine (SVM). The SVM algorithm is implemented using sequential minimal optimizing function (SMO). The features for the parser to machine learning include parts of speech, word and Lexicalized features. The algorithm achieved precision and recall of 82% for complex sentence parser and 89.5% for simple sentence. Accuracy of the result is 73.11%. The model created for the parser differs from the previous work since the model developed includes machine leaning technique and also the tag set used is different. At the end, the developed model gives satisfactory results