Neural Network

Neural Network

1. Abstract: A group of the IBM Research Laboratory in Brazil presented a paper titled Deep Convolutional Neural Networks for Sentiment Analysis of Short Texts, which proposes a deep convolutional neural network to explore from sentence-level character information to perform the analysis Of feelings of texts. The approach was applied in two different domains: the Stanford Sentiment Treebank (SSTb), which contains movie criticism phrases; And the Stanford Twitter Sentiment Corpus (STS), which contains Twitter messages. The proposed network, called Character to Sentence Convolutional Neural Network (CharSCNN), uses two convolutional layers to extract relevant characteristics of words and phrases of any size and also easily exploits the richness of embedded sentences produced by unsupervised pre-training. CharSCNN works as follows: given a sentence, a score is calculated for each feeling label. To mark a sentence, the network takes as input the sequence of words in the sentence and passes it