Twitter sentiment analysis using hybrid Spider Monkey optimization method

• Published in: Evolutionary intelligence Vol. 14; no. 3; pp. 1307 - 1316
• Main Authors: Shekhawat, Sayar Singh, Shringi, Sakshi, Sharma, Harish
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.09.2021; Springer Nature B.V
• Subjects: Applications of Mathematics; Artificial Intelligence; Bioinformatics; Cluster analysis; Clustering; Control; Data mining; Datasets; Digital media; Engineering; Evolutionary algorithms; Evolutionary computation; Heuristic methods; Mathematical and Computational Engineering; Mechatronics; Optimization; Robotics; Sentiment analysis; Social networks; Special Issue; Statistical Physics and Dynamical Systems; Vector quantization; Machine learning techniques; Nature-Inspired-Algorithm; Sentiment-Analysis (SA); Spider-Monkey optimizer; Twitter; Optimization; K-means clustering
• ISSN: 1864-5909; 1864-5917
• DOI: 10.1007/s12065-019-00334-2

The use of social media, over the past few years, has escalated enormously. Social media has formed a platform for the availability of abundant data. Thousands of people express their perceptions through social media. Sentiment Analysis (SA) of such views and perceptions is very substantial to measure public notion on a peculiar/specific subject matter of concern. SA is a remarkable field of data mining concerned with identification and translation of sentiments accessible on social media. Twitter is a microblogging site in which users can post updates (tweets) to friends (followers). This paper proposes a mechanism for extracting the sentiments from the tweets posted on Twitter. Tweets can be classified as positive, neutral or negative. The metaheuristic-based clustering techniques are superior to conventional techniques due to the subjective behaviour of tweets. A hybrid strategy, named as Hybrid Spider Monkey optimization with k-means clustering, is introduced to obtain the optimal cluster-heads of the dataset. The accuracy of the proposed method is determined on two datasets, namely, sender2 and twitter. To analyse the authenticity of the proposed method, a comparative analysis is performed with a few significant Nature-Inspired Algorithms such as Spider-Monkey optimization, Particle-Swarm algorithm, Genetic-Algorithm and Differential Evolution.