GL-LSTM Model for Multi Label Text Classification of Cardiovascular Disease Reports

EasyChair Preprint 7804

Abstract

In recent years, the rapid growth of electronic data and information has gotten a lot of attention, finding relevant information has become increasingly challenging. The goal of automatic text categorization is to classify textual articles based on their categories, especially in the medical domain. However, for some applica-tions, objects must inherently be described by more than one label. In this re-search, a new scheme of medical multi-label text classification is investigated which is based on intelligent engineering features using GloVe technique and LSTM classifier. The main particularity of GloVe consists of extracting informa-tive features to the word level automatically, allowing to capture global and local textual semantics. The choice of the LSTM model is motivated by the success that has been achieved in recent years; it allows to capture very long-term dependen-cies between words. The experiment of our approach named GL-LSTM on the cardiovascular text dataset has produced impressive results with an overall accu-racy of 0.927 compared with related work existing in the literature.

Keyphrases: Classification, GloVe, LSTM, medical text, multi-label, text categorization

@booklet{EasyChair:7804,
  author    = {Rim Chaib and Nabiha Azizi and Didier Schwab and Ibtissem Gasmi and Amira Chaib},
  title     = {GL-LSTM Model for Multi Label Text Classification of Cardiovascular Disease Reports},
  howpublished = {EasyChair Preprint 7804},
  year      = {EasyChair, 2022}}