Automatic keywords extraction from the domain texts: Implementation of the algorithm based on the MapReduce model

Aliya Nugumanova, Artem Novosselov, Yerzhan Baiburin, Alexey Karimov

2013

Closed access Metadata (CSL-JSON)
DOI 10.1109/ctit.2013.6749500
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Automatic keywords extraction is used in almost all the tasks related to natural language processing, such as annotation, indexing, classification, machine translation, knowledge extraction, etc. A large number of effective methods and approaches were developed to solve this problem, and the most simple and robust ones of them are based on the statistics of words. In this paper we describe a statistical method based on Chi-square test. The traditional algorithm implementing this method is an inefficient and time-consuming one. The aim of the paper is to develop the algorithm of this method based on distributed computing model. So we describe the implementation of the algorithm based on the MapReduce model of distributed computing and present the results of experiments showing the benefits of distributed computing.