Harshit Monish

Harshit Monish

Software Engineer (Deep Learning Compilers) at Nvidia Corporation || Ex-Machine Learning Intern at Intel || Ex-Google Developer Student Club President at UB || Ex-Lead Software Engineer at Cadence Design Systems

Publications

A Comparative assessment of Data Mining Algorithms to predict fraudulent firms

Published in IEEE International Conference on Cloud Computing, Data Science & Engineering(Confluence), 2019

In this paper, we compare the classification models of Data mining that can predict the fraudulent firm based on current and past risk factors. Further, we apply the ensemble techniques to improve the models and finally compare the models based on accuracy and compute complexity. The dataset collected is multivariate having 18 attributes that have been considered as risk factors.

Recommended citation: H. Monish and A. C. Pandey, "A Comparative Assessment of Data Mining Algorithms to Predict Fraudulent Firms," 2020 10th International Conference on Cloud Computing, Data Science & Engineering (Confluence), 2020, pp. 117-122, doi: 10.1109/Confluence47617.2020.9057968. https://ieeexplore.ieee.org/document/9057968

Intelligent Jobs Scheduling in palladium Cloud using Machine learning

Published in Cadence TeqFest 2019, 2019

In this poster we put forward a job scheduling algorithm with the goal of the minimum completion time, maximum load balancing degree using an improved deferential algorithm in cloud computing. Our proposed solution is to use machine learning (ML) techniques, especially supervised learning techniques to learn and predict the compile time of a design at the parser stage of the palladium frontend compiler.

Download here