Analysis of Parkinson’s Disease using Deep Learning and Word Embedding Models

• DOI: 10.33793/acperpro.02.03.86
• Publication Date: 2019-12-02T14:48:04Z
• Authors: Feyza Cevik; Zeynep Hilal Kilimci
• Keywords: Word embedding; Word2vec

Parkinson's disease is a common neurodegenerative neurological disorder, which affects the patient's quality of life, has significant social and economic effects, difficult to diagnose early due gradual appearance symptoms. Examining discussion Parkinson’s in media platforms such as Twitter provides platform where patients communicate each other both diagnosis treatment stage disease. The purpose this work evaluate compare sentiment analysis people about by using deep learning word embedding models. To best our knowledge, very first study analyze from models algorithms. In study, Word2Vec, GloVe, FastText are employed for enriching tweets terms semantic, context, syntax. Convolutional Neural Networks (CNNs), Recurrent (RNNs), Long Short-Term Memory (LSTMs) implemented classification task. This demonstrates efficiency algorithms understand needs patients’ provide valuable contribution process analyzing sentiments them with 93.63% accuracy performance.