Hadoop framework for efficient sentiment classification using trees. Issue 5 (22nd September 2020)

• Record Type: Journal Article
• Authors: Sridharan, K.; Komarasamy, G.; Daniel Madan Raja, S.
• Is Part Of: IET networks. Volume 9:Issue 5(2020)
• Page Start: 223
• Page End: 228
• Publication Date: 2020-09-22
• Subjects: feature extraction -- data handling -- pattern classification -- Big Data -- learning (artificial intelligence) -- data analysis -- sentiment analysis -- tree data structures; extensive data analysis -- opinion words -- sentiment analysis -- stipulated time limit -- traditional database software tool -- big data -- machine learning algorithms -- massive volume -- efficient sentiment classification -- Hadoop framework; Computer network architectures -- Periodicals; Computer network protocols -- Periodicals; Information networks -- Periodicals; Telecommunication systems -- Periodicals; 004.605
• DOI: 10.1049/iet-net.2019.0208
• Languages: English
• ISSNs: 2047-4954

Abstract

Abstract : Due to the increase in the speed of generation of data, the authors are forced to handle a massive volume of data with the help of conventional machine learning algorithms. Big data is an enormous volume of data which is beyond the capacity of the traditional database software tool to collect, store, manage, and process within a stipulated time limit. Sentiment analysis is analysing the data by classifying the text on the basis of strength and polarity of opinion (positive/negative) words that define the text. While handling big data, Hadoop provides a platform for users to develop their own sentiment analysis with the help of a lexicon dictionary or available application programming interface (API) or external programs. The aim of classifying data is to analyse extensive data and develop an appropriate description or model for every organised class with the feature present in the data. In this work, the feature extraction based on term frequency‐inverse document frequency is utilised and the Hadoop framework in attaining a useful classification with the help of random forest techniques.