Sentiment Mining of Online Reviews Using Machine Learning Algorithms

  Vidushi,  Gurjot Singh Sodhi

Social media has drawn the attention of researchers all around the world in recent times. The reason can be attributed to the large set of data available due to active involvement of the users on such platforms. The paper proposes a novel strategy of sentiment analysis on user’s review data using hybrid algorithm. Analysis of public’s views regarding a particular items is a complex task which involves many aspects like preprocessing, score calculation, classification algorithm etc. The paper proposes a novel strategy in which the effect of other’s review for score calculation are taken into account. Also the grammatical mistakes are taken into account for pre-processing. Further a hybrid KNN algorithm will be developed which will address the short comings of earlier used algorithms like SVM for handling high dimensionality data through chi-square technique. The results will be compared to that of traditional algorithms in terms of precision, accuracy and recall values.

Keywords- KNN,SVM, Naïve Bayes , social review, hybrid knn, data mining

Unique Identification Number - IJEDR1702208
Page Number(s) - 1321-1334
Pubished in - Volume 5 | Issue 2 | May 2017
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Vidushi,  Gurjot Singh Sodhi,   "Sentiment Mining of Online Reviews Using Machine Learning Algorithms", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.5, Issue 2, pp.1321-1334, May 2017, Available at :http://www.ijedr.org/papers/IJEDR1702208.pdf