Social media data on Web sites such as Twitter, Facebook, LinkedIn, YouTube, and Instagram is increasing tremendously because of their significant number of users. Newspaper, radio, television provide one-way communication, whereas social media provides many-to-many communication. Thus, analysis of social media data can produce many hidden information. Frequent patterns in social media can generate hidden information that can be useful. In this paper, we discussed how parallel frequent pattern mining algorithm is useful in finding patterns of natural language-based social media data. We present a process to retrieve frequent patterns (or rules) from a social media using thresholds of support and confidence. The parallel computing is achieved with the help of a scalable Apache Spark program. The retrieved patterns can be useful in making decisions related to social media.
CITATION STYLE
Chaturvedi, S., & Saritha, S. K. (2019). Parallel frequent pattern mining on natural language-based social media data. In Advances in Intelligent Systems and Computing (Vol. 814, pp. 507–517). Springer Verlag. https://doi.org/10.1007/978-981-13-1501-5_44
Mendeley helps you to discover research relevant for your work.