Open AccessA Swarm Intelligence Based Weighted Feature Extraction and Classification using SVM for Sentimental Exploration
Author(s) -
P. V. Naga Srinivas*,
M. V. P. Chandra Sekhara Rao
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.c4077.098319
Subject(s) - particle swarm optimization , computer science , support vector machine , artificial intelligence , data mining , set (abstract data type) , measure (data warehouse) , feature (linguistics) , machine learning , pattern recognition (psychology) , swarm behaviour , categorization , swarm intelligence , linguistics , philosophy , programming language
The goal of Sentiment Exploration (SE) is used for mining the accurate sentiments which are very beneficial for businesses, governments, and individuals, the opinions, recommendations, ratings, and feedbacks are becoming an important aspect in present scenarios. The proposed methodology likewise attempts to introduce a swarm intelligence based sentimental supervised methodology. In order to obtain a relevant feature data set from a large number of data samples, this method used particle swarm optimization to attain the utmost optimum feature set. The evaluation of the optimum feature set is obtained by means of using Minimum Redundancy and Maximum Relevancy measure as the fitness function. The categorization of the extracted feature set is accomplished with the Support Vector Machine classification technique. The experimental outcome for the suggested method is evaluated using four performance measure like precision, recall, accuracy, and f-measure and showed that proposed swarm intelligent based classification method has better performance using IMDB, Movie Lens and Trip Advisor Data Samples.