• Mahesh S Nayak
• Dr. M. Hanumanthappa
• Dr. S.Kavitha

Big data is gathering and analyzing the data. Without analytics, it’s just a bunch of data with limited business use. The big amounts of data storing from various sources the significance of analytics has enormously grown making the companies to tap the bunch of data that was considered useless all these years. The importance of big data is given preference as bound to provide results on the fly. The MapReduce paradigm has long been a staple of big data computational strategies. With the development of the Internet with website, people communicate online news articles every day. The percentage of message communication any news article indicates how popular the news is. The objective of the paper is to find the best model and set of feature to predict the popularity of online news, using machine learning techniques. The dataset is collected from UCI Mashable online news website. For the feature set preprocessing is done using unsupervised learning method AddExpression-E 0.0 expression algorithm. We have implemented two classififiers using ZeroR and RepTree classification. We have successfully implemented two clustering algorithms KMeans clustering and Canopy cluster algorithm. Their performances are recorded and compared. The ZeroR gives the best result with less time complexity of 0.05 seconds and canopy clustering took 2.09 seconds. The research can be used in any organization to choose the classifier and cluster algorithm for process the dataset.

Big Data, Web, Machine learning; Classification; Clusters

"Big Data and Web : DISC in Web Analytics For Large Data Popularity of Online News on Mapreduce Clusters", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 7, page no.336-344, July-2018, Available :http://www.jetir.org/papers/JETIR1807054.pdf

"Big Data and Web : DISC in Web Analytics For Large Data Popularity of Online News on Mapreduce Clusters", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 7, page no. pp336-344, July-2018, Available at : http://www.jetir.org/papers/JETIR1807054.pdf