An Improved Hidden Markov Model for Literature Met

Authors: BinGe Cui Xin Chen

Publish Date: 2010/8/18

Volume: , Issue: , Pages: 205-212

PDF Link

Abstract

In this paper we proposed an improved Hidden Markov Model HMM to extract metadata in the academic literatures We have built a dataset including 458 literatures from the VLDB conferences which contains the visual feature of text blocks Our approach is based on the assumption that the text blocks in the same line have the same state information type The assumption is effective in more than 98 occasions Thus the state transition probability among the same states in the same line is much larger than that in different lines According to this conclusion we add one state transition matrix for HMM and modified the Viterbi algorithm The experiments show that our extraction accuracy is superior to that of any existing works

Keywords: