Automated sarcasm detection and classification using hyperparameter tuned deep learning model for social networks. Issue 10 (22nd July 2022)

• Record Type: Journal Article
• Authors: Vinoth, Dakshnamoorthy; Prabhavathy, Panneer
• Other Names: Dhiman Gaurav guestEditor.; Nagar Atulya K. guestEditor.; Gupta Deepak guestEditor.
• Is Part Of: Expert systems. Volume 39:Issue 10(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-07-22
• Subjects: classification -- deep learning -- hyperparameter tuning -- sarcasm detection -- social networks; Expert systems (Computer science); 006.33
• DOI: 10.1111/exsy.13107
• Languages: English
• ISSNs: 0266-4720

Abstract

Abstract: In recent digital era, social media sites have been commonly used by majority of people to generate massive quantities of textual data. Sarcasm can be treated as a kind of sentiment, which generally expresses the opposite of what has been anticipated. Since sarcasm detection is mainly based on the context of utterances or sentences, it is hard to design a model to proficiently detect sarcasm in the domain of natural language processing (NLP). The recent advancements of deep learning (DL) models influence neural networks (NN) in learning the lexical as well as contextual features, eradicating the necessity of hand‐crafted features for sarcasm detection. With this motivation, this article designs an automated sarcasm detection and classification tool using hyperparameter tuned deep learning (ASDC‐HPTDL) model for social media. The proposed ASDC‐HPTDL technique primarily involves pre‐processing stage to transform the data into useful format. At the next stage of pre‐processing, the pre‐processed data is converted into the feature vector by Glove Embedding's technique. Followed by, attention bidirectional gated recurrent unit (ABiGRU) technique is utilized to detect and classify sarcasm. In order to boost the detection outcomes of the ABiGRU technique, a hyperparameter tuning process using improved artificial flora algorithm (IAFO) is employed, shows the novelty of the work. The proposed model is validated using the benchmark dataset and the results are examined interms of precision, recall, accuracy, and F1‐score.