Author : S.N.Sithi Shamila 1
Date of Publication :27th December 2017
Abstract: The revolution that has swept through information Processing arena has placed the modern world in a unique vantage position. Huge quantum of data has been generated through the normal operations of day-to-day business activities in various fields such as public services, Research, Education, Industries and Institutes. The multiple level transactions captured in these fields have generated such an enormous amount of data which are so challenging to handle as it swings from megabytes to petabytes, and in a swift, to zettabytes as well. The inevitability of the dependence on the Internet in this technologically revolutionized context has paved way for the increased use of data, and this situation has obviously generated considerable research interest to study user view analysis in order to determine the several characteristics of the users which might influence the development of software systems and several other technological products. It has been observed that more than 90 percent of today’s data is either unstructured or semi-structured, and this has complicated and challenged the study of decoding knowledge and information embedded in different recognizable patterns and analyze the text documents from the huge volume of accumulated data. It is in this context, the study of text mining, which is essentially a process of extracting interesting and non-trivial patterns from a large number of text documents, assumes significance. Although there are innumerable techniques and tools to mine the text in order to discover valuable information which with not only future predictions in several areas of business, social, political and economic interest can be made, also it will influence the process of decision making in several spheres. Thus, Sentiment Analysis or Opinion Mining has attained an enormous importance in the modern world today, and it is in this context, an attempt has been made in this paper to briefly discuss and analyze text mining techniques and the issues connected to text mining which affect accuracy and relevance of the results obtained.
Reference :
-
- 1. Diana Maynard, Adam Funk. Automatic detection of political opinions in tweets. In: Proceedings of the 8th international conference on the semantic web, ESWC‘11; 2011
- 2. Hastie, T., Tibshirani, R., & Friedman, J. (2009). The Elements of Statistical Learning: Data Mining, Inference, and Prediction.Springer.
- 3. Hu Minging, Liu Bing. Mining and summarizing customerreviews. In: Proceedings of ACM SIGKDD international conference on Knowledge Discovery and Data Mining (KDD‘04);2004.
- 4. Liu B. Sentiment analysis and opinion mining. Synth Lecture Human Language Technology 2012.
- 5. Machine Learning with Naïve Bayes Classifier [Online].Available:http://blog.datumbox.com/mach ine-learningtutorial-the naive-bayes-text-classifier/
- 6. TsytsarauMikalai, Palpanas Themis. Survey on mining subjective data on the web. Data Mining Knowledge Discovery.
- 7. Wilson T, Wiebe J, Hoffman P. Recognizing contextual polarity in phrase-level sentiment analysis. In: Proceedings of HLT/EMNLP; 2005
- 8. Yelena Mejova, Padmini Srinivasan. Exploring feature definitionand selection for sentiment classifiers. In: Proceedings of the fifthinternational AAAI conference on weblogs and social media;2011.