Author(s): Mr. Amin Nazir Nagiwale, Mr. Manish R. Umale, Mr. Aditya Kumar Sinha
Published in: International Journal of Engineering Research & Technology
License: This work is licensed under a Creative Commons Attribution 4.0 International License.
Volume/Issue: Volume. 5 - Issue. 02 , February - 2016
Social Media is emerged as an open platform that people use it to share/exchange information. Data captured from social media is generated in different formats, different sources, varies in real time i.e. data generated has big data properties like velocity, volume, variety. For class of machine learning algorithms, large scale data processing frameworks like Hadoop, MapReduce are insufficient as they requires intermediate results across multiple executions. Also they have to dealtwith skewness, diversity of data to adapt changes in real time. For example Social media site, Twitter generates approximately 6000 tweets per second and these are according to trends that changes over time. As success of any machine learning algorithm like Sentiment Analysis is depend upon how well it is trained on training set, domain-specific information and corpus of training set. To overcome these problems we have proposed Selfie,Self- adjusting algorithm in .In this paper we have discussed preparing domain –specific corpus according to change in trends in real time by using self-adjusting computation. We have implemented Splay tree with trends as tree node and trends near to root of the tree are the most talked/tweeted trends. Each trend node contributes towards domain–specific corpus according to height of tree.
Number of Citations for this article: Data not Available
7 Paper(s) Found related to your topic: