A study on high dimensional big data using predictive data analytics model

Nivethitha Krishnadoss, Lokesh Kumar Ramasamy

Abstract


A massive bulk of data is being created due to digitalisation in various industries, including medical, manufacturing, sales, internet of things (IoT) devices, the web, and businesses. To find data patterns for data attributes machine learning (ML) algorithms are used. In this fast-growing world, we can see that data is generated in abundance by people, machines, and corporations. With the increase in computer science market, researchers are integrating heterogeneous and diverse data into accurate patterns by applying machine learning algorithms and complex strategies on data sets. The overabundance of high-dimensional big data has made it more difficult for scientists to extract important information from these data efficiently. Conventional data mining approaches are ineffective when dealing with large amounts of data. As big data increase exponentially, predictive analytics has become widely known. To evaluate a large number of data patterns, data driven technology predictive big data analytics (PBA) can be used and ML algorithms to investigate the present and future data based on the records of data patterns. In this research paper, predictive analysis on big data has been proposed using the splitting random forest (SRF) methodology with help of hyperparameter optimization and dimension reduction technique.

Keywords


Dimension reduction; Hyperparameter optimization; Machine learning; Predictive big data analytics; Splitting random forest

Full Text:

PDF


DOI: http://doi.org/10.11591/ijeecs.v30.i1.pp174-182

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics