Rohit Kumar Kaliyar, K. Ram, Akansha Sharma, Smita Tiwari, N. Ahuja, Mohit Agrawal
{"title":"Affects in Tweets with Real Time Emotions using Deep Learning Techniques: A Novel Approach","authors":"Rohit Kumar Kaliyar, K. Ram, Akansha Sharma, Smita Tiwari, N. Ahuja, Mohit Agrawal","doi":"10.1109/Confluence47617.2020.9057913","DOIUrl":null,"url":null,"abstract":"Twitter is an online microblogging tool that has 400 million messages per day. SemEval-2018 Tasks have already been presented and explored in the previous years by the name of “Affect in Tweets” but the scope for improvement never ends. So, in this research paper, we come up with deep learning architecture which is extremely coherent for the given task of extracting emotion intensity and classes from tweets (description of the task is given on www.codalab.com for details). Deep learning models are productive due to their automatic learning capability and automatic feature extraction. This research paper highlights the implementation of deep learning-based models such as convolutional neural networks and LSTM for classifications. The implemented tasks are-:1. emotion intensity regression 2. Emotion intensity ordinal classification,z 3. Multilabel emotion classification. We have expressed that the fine-grained intensity scores that we have obtained are reliable. Our dataset is beneficial for testing supervised machine learning algorithms for multi-label classification, intensity regression, sleuthing ordinal category of intensity of feeling (low, moderate, etc.). We have implemented various machine learning and deep learning-based models and achieved an accuracy of 77.64% in E-oc (Emotion ordinal classification) task, which is the highest among all competitors.","PeriodicalId":180005,"journal":{"name":"2020 10th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 10th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/Confluence47617.2020.9057913","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Twitter is an online microblogging tool that has 400 million messages per day. SemEval-2018 Tasks have already been presented and explored in the previous years by the name of “Affect in Tweets” but the scope for improvement never ends. So, in this research paper, we come up with deep learning architecture which is extremely coherent for the given task of extracting emotion intensity and classes from tweets (description of the task is given on www.codalab.com for details). Deep learning models are productive due to their automatic learning capability and automatic feature extraction. This research paper highlights the implementation of deep learning-based models such as convolutional neural networks and LSTM for classifications. The implemented tasks are-:1. emotion intensity regression 2. Emotion intensity ordinal classification,z 3. Multilabel emotion classification. We have expressed that the fine-grained intensity scores that we have obtained are reliable. Our dataset is beneficial for testing supervised machine learning algorithms for multi-label classification, intensity regression, sleuthing ordinal category of intensity of feeling (low, moderate, etc.). We have implemented various machine learning and deep learning-based models and achieved an accuracy of 77.64% in E-oc (Emotion ordinal classification) task, which is the highest among all competitors.