An Analysis of Variance Method for Detection of Collocations in a Pedagogical Domain Corpus

Abstract

In this paper, an exploratory experiment, based on analysis of variance, was carried out in order to get collocations in a pedagogical domain corpus. A semi-automatic corpus containing learning styles papers in Spanish was built. After wards, the corpus was lemmatized and a bigrams representation was extracted. The proposed method consists on divide the list of bigrams in quartiles, and analyzing the variance on each one of them. A list of collocations, which was evaluated using a gold standard built by an expert in the domain, was retrieved from each experiment accordingto established thre sholds for the method. Results showed a retrieved list with important collocation in the selected domain.