Automatic classification of Lithuanian Parliament bills

Abstract

Computer systems are omnipresent nowadays. Quantitative techniques are getting more and more important in political science. Therefore, computer technologies are required for political data processing and (semi) automatic analysis. In this paper we apply Natural Language Technologies, mainly text classification, to categorize bills of Lithuanian parliament into the predefined groups, for further use in voting analysis and other political science tasks. Only titles of bills were used, therefore, it generalizes to a short text classification problem, which is poorly explored for Lithuanian language