Open AccessSentimental Analysis on Text data by using Unsupervised Methods
Author(s) -
B.Manjula Josephine,
K. R. Rao,
K. Ramya,
P. Sandeepa,
Gunda Sai Yeshwanth
Publication year - 2019
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.a9385.109119
Subject(s) - computer science , product (mathematics) , cluster analysis , identifier , parsing , representation (politics) , information retrieval , face (sociological concept) , noun , natural language processing , data science , artificial intelligence , linguistics , programming language , philosophy , geometry , mathematics , politics , political science , law
On the internet we can see how efficiently display the reviews by the user who brought the product so that it covers all the important points instead of just displaying few top comments or threads. The main agenda of the tool is to build and analyse all the reviews given by each customer and display the best product reviews for any app or product. As we all read reviews before we buy any product from any e-commerce or while installing any app but the major problem we face is there are huge number of reviews and most of the reviews we get is the top most review or a combination of bad and good review based on rating which sometimes may or may not tell the perks or cons of using the product so we tried to build a tool that analyse all the reviews and picks the best reviews which totally describe the product flaws defects or advantages. so for that purpose we are implementing the k-means clustering algorithm and in previous papers they have used RASP (robustical and accurate statistical parser)grammatical tagger to identify all kinds of nouns, adjectives, pronouns and etc. together. Here in our paper we are using k-means clustering algorithm which divides all kinds of identifiers based on the comments so that it gives a clear idea about the product and is given in graphical representation