Krishnendra Sankar Ganguly

Work place: Ernst & Young GDS, Kolkata, India

E-mail: krish.ksg@gmail.com

Website:

Research Interests:

Biography

Krishnendra Sankar Ganguly holds Bachelor’s degree in Science in Computer Science from Sam Higginbottom Institute of Agriculture Technology & Sciences (SHIATS), U.P., prior to completion of his three Master’s degrees (i) in Computer Science from SHIATS, U.P., (ii) in Computer Application from National Institute of Electronics & Information Technology, Govt of India and (iii) in Technology in Computer Science & Engineering from Birla Institute of Technology, Mesra. He later specialized his knowledge in Data Science, Machine Learning & Deep Learning, Artificial Intelligence & Neural Networks, Predictive Analysis & Forecasting, etc., particularly in Inter-disciplinary Research in Bio-Medical arena. For the last few years, he has been rendering his services in the post of Senior Software Engineer in globally-renowned Enterprises, viz, WIPRO and Ernst & Young GDS.

Author Articles
A Comparative Study of Statistical (SARIMA) Vis-À-Vis Some Traditional Machine-Learning and Deep-Learning Techniques to Forecast Malaria Incidences in Kolkata of India

By Krishnendra Sankar Ganguly Krishna Sankar Ganguly Ambar Dutta

DOI: https://doi.org/10.5815/ijitcs.2025.05.06, Pub. Date: 8 Oct. 2025

To augment the accuracy of the results of a Time-Series Forecasting problem in the Computational Epidemiology domain of Public Health, to generate an accurate alert in a Real-time Outbreak and Disease Surveillance (RODS) system, namely in the prediction of Malaria incidences, an interdisciplinary approach of data analysis [through Statistical along with Machine-Learning (ML) and Deep-Learning techniques (DL)] has been studied in this research. Two different Non-linear Deep-Learning based techniques, viz., Long Short-Term Memory (LSTM) [a subclass of Recurrent Neural Network (RNN)] & Gated Recurrent Unit (GRU) and two different Non-linear Machine-Learning techniques, viz., Random Forest Regressor & Non-linear Support Vector Machine Regressor are applied in this study to compare against the traditional Statistical-based linear SARIMA model, to forecast a longitudinal data-set of malaria incidences. While SARIMA or other traditional Autoregressive (AR) models, necessitating a smaller number of parameters, undergo limited training and limited prediction power, ML and DL models show profound and persistent performance improvement with better noise-handling/ missing values and perform multi-step forecasts. Moreover, the over-fitting issue can be combated by introducing densely connected residual links in the ML/ DL networks.

[...] Read more.
Other Articles