An Intelligent Deep Neural Sentiment Classification Network

A Deep Neural Sentiment Classification Network (DNSCN) is developed in this work to classify the Twitter data unambiguously. It attempts extract negative and positive sentiments database. The main goal of system find sentiment behavior tweets with minimum ambiguity. well-defined neural network extracts deep features from automatically. Before extracting deeper deeper, text each tweet represented by Bag-of-Words (BoW) Word Embeddings (WE) models. effectiveness DNSCN architecture analyzed using Twitter-Sanders-Apple2 (TSA2), Twitter-Sanders-Apple3 (TSA3), Twitter-DataSet (TDS). TSA2 TDS consist tweets, whereas TSA3 has neutral also. Thus, proposed acts as a binary classifier for databases multiclass TSA3. performances are evaluated F1 score, precision, recall rates 5-fold 10-fold cross-validation. Results show that DNSCN-WE model provides more accuracy than DNSCN-BoW representing feature encoding. score DNSCN-BW based on database 0.98 (binary classification) 0.97 (three-class This better 0.99