A SURVEY OF ADAPTIVE TASK SCHEDULING FOR SOCIAL BIG DATA
Keywords:
Big Data, Social Media Data, Text Analytics, Network Analysis, Predictive Modeling, Information Diffusion, Information Fusion, Apache Hadoop, Apache SparkAbstract
Social media is a large source of Big Data. Its data is continuously increasing and changing therefore it requires new and innovative forms of information processing. Research areas like Data Mining, Machine learning and Social Networks finds its application in analyses of Big Data. Different graph and networks processing algorithms like calculating centrality, identifying clusters and identifying sources of information diffusion are applied on user graphs. Most of the data is in unstructured format of texts. This gives rise to perform different Text Analytics methodologies on social media data. Variety of tools, machine learning libraries and frameworks are developed for effective utilization of Data Mining Methods. Innovative methods of data management are also required as traditional storages are ineffective for unstructured data. These Big Data processing methods found their applications in wide areas like Marketing, Criminal activities and Fraud detection, Epidemic Intelligence etc. There are also a number of open challenges in these
processing techniques. This paper gives an overview of methodologies and frameworks for processing social media big data.
