Multiclass Event Classification from Text

Social media has become one of the most popular sources information. People communicate with each other and share their ideas, commenting on global issues events in a multilingual environment. While social been for several years, recently, it given an exponential rise online data volumes because increasing popularity local languages web. This allows researchers NLP community to exploit richness different while overcoming challenges posed by these languages. Urdu is also used being media. In this paper, we presented first-ever event detection approach language text. Multiclass classification performed deep learning (DL) models, i.e.,Convolution Neural Network (CNN), Recurrence (RNN), Deep (DNN). The one-hot-encoding, word embedding, term-frequency inverse document frequency- (TF-IDF-) based feature vectors are evaluate Learning(DL) models. dataset that experimental work consists more than 0.15 million (103965) labeled sentences. DNN classifier achieved promising accuracy 84% extracting classifying script.