Subanta pada Analyzer for Sanskrit

Natural language processing has wide coverage in application areas like machine translation, text to speech conversion, semantic analysis, semantic role labeling and knowledge representation. Morphological and syntactic processing are components of NLP which process each word to produce the syntactic structure of the sentence, with respect to its grammar. Semantic analysis follows syntactic analysis. One of the key task in morphological analysis is identifying the correct ischolar_main word from its inflected form. In Sanskrit language, these inflected word follow the rules which are used to separate the ischolar_main word from its suffix. These extracted suffix carry sufficient amount of syntactic and semantic information with them. To develop such word splitter, rules called sandhi rules given in the grammar of the Sanskrit language has been used. The challenge in the problem lies is in identifying the junction point or breaking point of the word as multiple junction can be obtained within the word. Developed system maintains database of all possible suffix, which are then used for splitting the word. The algorithm for the same is presented in the paper with the solutions to problem faced while developing the module.