PHD Research Proposal in Predictive Analytics

Predictive analytics is one of the subdivisions of advanced analytics that seeks to predict future events by reference to historical events. With the advent of disparate machine learning methods and statistical methods, predictive analytics system significantly captures the potential insights from the large volume data chunks for predicting the behavior patterns and future trends [1]. In order to improve the bottom line, optimize the processes, discover opportunities and risks and guide decision making, most of the organization relies on predictive analytics
[2]. It is the crucial data-driven methodology to recognize the concealed pattern from the unprecedented data for predicting probabilities of forthcoming according to the past occurrence knowledge. By leveraging the predictive analytics on the big data brings the tangible business benefit to the organization in which predictive analytics enable the in-depth insight of customer behavior. The predictive analytics shows their supremacy of performance in various fields include banking sector, fraud detection, risk management system, health care, dynamic product pricing, sales forecasting, weather forecasting, and direct marketing
[3].Most of the organizations seek to take advantage of predictive analytics to gain significant benefits through the faster identification of risks and opportunities. Even though, they are lacking in some cases. The significant one is the inadequacy of up-to-date data that makes the severe impacts on the efficiency of the results. Some enterprises have scientific experts to overcome the circumstance of predictive analytics and to make an accurate prediction. However, the lack of resources includes budget intends to degrade the accuracy of the system.
[3].The fusion of predictive analytics with big data brings valuable insights from the large-scale data. However, it often suffers from the issues while getting the real, adequate and perfect data to discover the efficiency of its model. Moreover, some of the prediction models do not bring the relevant results owing to the lack of selection knowledge of the variables and algorithm that makes the negative impact on the accuracy of prediction. In business applications, predictive analytics need to incorporate efficient data management strategies to integrate and standardize the incoming heterogeneous data. The lack of efficient data management misleads the prediction.

Reference:

  • [1] Kotu, Vijay, and Bala Deshpande, ”Predictive analytics and data mining: concepts and practice with rapidminer”, Morgan Kaufmann, 2014.
    [2] Shmueli, Galit, and Otto R. Koppius, “Predictive analytics in information systems research”, MIS quarterly, pp.553-572, 2011.
    [3] Watson, Hugh J, “Tutorial: Big data analytics: Concepts, technologies, and applications”, CAIS, Vol.34, pp.65, 2014..

  • [2]Kaur, Tarandeep, and Inderveer Chana, “Energy efficiency techniques in cloud computing: A survey and taxonomy.” ACM Computing Surveys (CSUR), Vol.48, No.2, pp.22, 2015.

  • [3] Kansal, Nidhi Jain, and Inderveer Chana, “An empirical evaluation of energy-aware load balancing technique for cloud data center,” Cluster Computing, Vol.21, No.2, pp.1311-1329, 2018.

  • [4] Guo, Songtao, Bin Xiao, Yuanyuan Yang, and Yang Yang, “Energy-efficient dynamic offloading and resource scheduling in mobile cloud computing,” 35th Annual IEEE International Conference on INFOCOM -The Computer Communications, IEEE, pp.1-9, 2016.

  • [5] Zhou, Zhou, Jemal Abawajy, Morshed Chowdhury, Zhigang Hu, Keqin Li, Hongbing Cheng, Abdulhameed A. Alelaiwi, and Fangmin Li, “Minimizing SLA violation and power consumption in Cloud data centers using adaptive energy-aware algorithms,” Future Generation Computer Systems, Vol.86, pp.836-850, 2018.

Leave Comment

Your email address will not be published. Required fields are marked *

clear formSubmit