TANA: The amalgam neural architecture for sarcasm detection in indian indigenous language combining LSTM and SVM with word-emoji embeddings

Sentiment analysis is indeed a difficult task owing to the playful language mannerism, altered vocabulary and speak-text used on online forums. Humans tend use words phrases in ways that are incomprehensible those who not involved discourse. Sarcastic remarks conversations often utilized mock others by saying something isn't pleasant. Sardonic or humorous statements/ tones insult make appear puerile. Automated sarcasm detection considered as one of key tasks tweak sentiment extending it morphologically rich free-order dominant indigenous Indian Hindi another challenge. This research puts forward ‘The Amalgam Neural Architecture’, TANA, detect tweets. The architecture trained using two embeddings, namely word emoji embeddings combines an LSTM with loss function SVM for detection. We Sarc-H dataset, which built scrapping tweets manually annotating based hashtags ‘’ (pronounced kataaksh, means Hindi) vyangya, tweeters results evaluated various classification performance metrics achieves F-score 0.9675 outperforming last layer softmax well existing works.