Publication Type : Conference Paper
Publisher : International Conference on Advances in Computing and Data Sciences
Source : International Conference on Advances in Computing and Data Sciences, (2021) Communications in Computer and Information Science, 1441, pp. 42-51., DOI: 10.1007/978-3-030-88244-0_5
Url : https://link.springer.com/chapter/10.1007/978-3-030-88244-0_5
Campus : Coimbatore
School : School of Artificial Intelligence, School of Artificial Intelligence - Coimbatore
Center : Center for Computational Engineering and Networking
Year : 2021
Abstract : This paper is an attempt to show a dependency parser for Hindi Language using Integer Linear Programming. A new approach is described for developing the constraint parser by convexifying the integer values taken as input, depicting the source groups (nouns) and demand groups (verbs) in a sentence. The convexified input is further used to find the Integer dependency values, which validates the word’s dependency relations in any given Hindi Language sentence. Based on the Karaka constraint conditions, Karaka charts are computed by referencing the Karaka word values obtained from the constraint formulations, which solve the inequalities generated in a sentence from its source groups and demand groups based on the concepts of Paninian Grammar.
Cite this Research Publication : Kesav, R.S., Premjith, B., Soman, K.P., "Dependency Parser for Hindi Using Integer Linear Programming," International Conference on Advances in Computing and Data Sciences, (2021) Communications in Computer and Information Science, 1441, pp. 42-51., DOI: 10.1007/978-3-030-88244-0_5