Development of email classifier in Brazilian Portuguese using feature selection for automatic response

TL;DR

This paper develops a Brazilian Portuguese email classifier using feature selection and compares different text preprocessing techniques, achieving up to 87.3% accuracy with SVM and POS filtering.

Contribution

It introduces a novel corpus for Brazilian Portuguese email classification and evaluates the impact of lemmatization and POS filtering on classifier performance.

Findings

SVM with POS filtering achieved 87.3% accuracy.

Lemmatization reduced classification precision and recall.

POS filtering improved overall classification results.

Abstract

Automatic email categorization is an important application of text classification. We study the automatic reply of email business messages in Brazilian Portuguese. We present a novel corpus containing messages from a real application, and baseline categorization experiments using Naive Bayes and support Vector Machines. We then discuss the effect of lemmatization and the role of part-of-speech tagging filtering on precision and recall. Support Vector Machines classification coupled with nonlemmatized selection of verbs, nouns and adjectives was the best approach, with 87.3% maximum accuracy. Straightforward lemmatization in Portuguese led to the lowest classification results in the group, with 85.3% and 81.7% precision in SVM and Naive Bayes respectively. Thus, while lemmatization reduced precision and recall, part-of-speech filtering improved overall results.