A Comparative Study of Emotion Mining on Social Media Data

Yousuf, Iftekhar Uddin; Raju, Abir Ahamed; Abdur Rakib, Mohammad

DSpace Home
→
Faculty of Science and Information Technology
→
DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING
→
Project Report
→
View Item

dc.contributor.author	Yousuf, Iftekhar Uddin
dc.contributor.author	Raju, Abir Ahamed
dc.contributor.author	Abdur Rakib, Mohammad
dc.date.accessioned	2020-10-05T14:00:22Z
dc.date.available	2020-10-05T14:00:22Z
dc.date.issued	2019-04-25
dc.identifier.uri	http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/4479
dc.description	Emotions are a part of parcel of human nature that can be considered as inherited. Also it has been found that different human being expression of a particular emotion is uniform. Emotion mining is the method of detecting, analyzing human’s felling towards different proceedings, services, or any other particular interest. The most interesting and trending topic for the research field is emotion mining was started since the 20th century. It is also known as the detection or finding the opinions from the positive text and negative text. Besides, this era of web 2.0 end user Produced data over the Internet has prolonged more and more rapidly. Social Media platform and commercial website. For instance, Facebook, Instagram, Twitter, LinkedIn, Amazon, IMDB offer a platform to share their experiences, knowledge and views on the recent trend of politics, economics and other global- critical issue. Emotion mining accumulates online documents ranging from twitters, Facebook; product reviews blogs and other social media platform. As we know, Human decision making is always influenced by others thinking, ideas and opinions. While making any purchase online consumer usually checks opinions of others about the product. Emotion mining is the automated mining of attitudes, opinions, and emotions from text, speech, and database sources through Natural Language Processing (NLP). Emotional mining implicates classifying opinions in a text into categories like "positive" or ‘negative’ or "neutral". It’s often mentioned as subjectivity analysis, Sentiment analysis and appraisal extraction. Sentiments or Opinions contain public generated content about products, services, policies and politics. Customers have a habit of trust the opinion of other consumers, particularly those with past experience of a product or service, rather than company marketing. Social Media are influencing customer preferences by understanding their approaches and behaviors. A lot of research work is being done in the field of emotional mining due to its importance in the marketing level competition and the changing needs of the people. It requires the usage of a training set for its performance. We present a table summarizing all the studied work. In this comparative study, we have taken a systematic literature review process to identify areas well focused by researchers, least addressed areas are also emphasized giving an opportunity to researchers for ©Daffodil International University 3 further work, weakness, strengths, threats and opportunities, whose factors represent, in a sense, future work to be carried out. We have also tried to identify most and least commonly used feature selection techniques to find research gaps for future work. Finally, the emotion mining schemes have been compared in terms of accuracy with respect to some algorithms. Thus this paperwork provides a detailed analysis of the recent emotion mining schemes and throws light on new avenues for future research work in this area.	en_US
dc.description.abstract	Sentiment analysis or emotion mining is the method of computational and natural language processing based techniques which are used to extract, identify or characterize subjective information, such as opinions, comment that expressed in a given piece of text. The main contributions of this paper involve the sophisticated categorizations of the trend of research on sentiment analysis and its allied areas as well as to identify the most and least commonly used feature selection techniques to find out research gaps for future research. This paperwork gives a detailed analysis of the recent sentiment analysis schemes and directed towards new avenues for future research work in this region.	en_US
dc.language.iso	en	en_US
dc.publisher	Daffodil International University	en_US
dc.subject	Data Mining	en_US
dc.subject	Social Media	en_US
dc.subject	Emotion Recognition	en_US
dc.title	A Comparative Study of Emotion Mining on Social Media Data	en_US
dc.type	Other	en_US