Social network comment classification using fuzzy based classifier technique
Varsha Bairagi, Namrata Tapaswi · 2016
Now in these days use of social networking websites are growing different kinds and natures of user are being utilize this services frequently. In this content the anomalous user can misbehave or can be involved in unsocial activity over the clean environment. Furthermore people frequently use symbolic forms to convert other person by which pattern of unsolicited can be recognized by analysis these content. Therefore in this research paper we are going to investigate the different technique and approaches that recently developed by different researchers. Finally we proposed a model by associating this method for this conversion. Data mining and machine learning, offers to evaluate the data automatically for different applications. In order to perform such task the classification and clustering techniques are used. The classification of data is a supervised learning process thus the technique needs a set of attributes (patterns) and the associated class labels. The algorithm first prepare the mathematical model using the previous patterns of training sets and further these models are used to evaluate the test sets. Thus a two different classifiers namely Bayesian classifier and k-nearest neighbour algorithm is studied. Moreover, some of authors are suggested to implement the fuzzy classification technique for achieving the high accurate results. Therefore a new method using the fuzzy concept is proposed and implemented. Additionally the classical text classification models are used for comparative performance study. The comparative performance study shows the effectiveness of the proposed classification technique and able to produce the more accurate results as compared to traditional classifiers.