Project: National Mission for Sustaining the Himalayan Ecosystem (NMSHE)
Task Force 3: Forest Resources and Plant Diversity (Phase II).
Funded by: Department of Science & Technology, Ministry of Science & Technology, (GoI)
Description: To investigate winter, monsoon and extreme rainfall events using observations and modelling for microstructure of the event. To continuously measure ecosystem exchange and micro-met parameters from an oak dominated vegetations of central Himalaya. Performed data analysis for atmospheric and meteorological data. Prepared data for regular reports and data analysis.
Project: Investigation of rainfall vertical structure and rainfall induced erosivity over a Garhwal Himalayan station using in-situ observation and modeling
Funded by: DST-SERB (GoI)
Description: To investigate vertical profiles of rainfall and integral rainfall parameters during monsoon seasons using an in-situ Micro Rain Radar and Disdrometer measurement. To establish a relationship between the rainfall, CAPE, CIN and Soil moisture for Land-Atmospheric interaction. To establish a relationship between the rainfall intensity and kinetic energy for deduction of erosivity. To assess performance of WRF-ARW cloud microphysical schemes with respect to in-situ observations on simulating vertical profiles of integral rainfall parameter/s during selected rainfall events of monsoon season. Atmospheric and Meteorological data analysis and visualization is using MATLAB and Python with different statistical skills.
Dissertation Title: Detecting the YouTube videos having Fake View using Machine Learning
Description: Fake views problems are increasing day by day and becoming a large issue for any social media platform. In this research, we used machine learning algorithms using input from many sources of data present on videos, for example, the number of views, the number of likes, the number of dislikes, the number of comments, etc., but for detecting fake views. We used YouTube videos that contain spam in the comments for the classification of the dataset. With the help of these parameters, which are publicly available, we try to train a machine learning model to detect YouTube videos with fake views. After training, a test dataset is used by the machine learning model to check the accuracy of the model using scikit-learn, a Python library. In this research, we proposed a machine learning model to detect YouTube videos that have fake views