Machine Learning, Deep Learning, Data Science

S H R E Y A N S H G U P T A

CONTACT: + **-* * ***07719 EMAIL G I THUB L I NKEDIN S TACKEXCHANGE EDUCATION

B.Tech. Civil Engineering 2018-2022 DELHI TECHNOLOGICAL UNIVERSITY 7.57 CGPA (Till 5th Sem) CBSE (Class XII) 2018 SOMERVILLE SCHOOL, DELHI PCM: 90% AGG: 86%

CBSE (Class X) 2016 SOMERVILLE SCHOOL, DELHI 9.4 CGPA EXPERIENCE

MACHINE LEARNING RESEARCHER Calibre DTU, Delhi Ongoing CALIBRE (Computational Analytic Learning and Intelligence Based Research) is an Official Research Group of DTU, under the Department of Computer Engineering

Developed a feature selection algorithm using advanced soft-computing techniques to optimize the solution to fake news detection and hate speech detection problem.

Currently working on COVID-19 risk assessment and travel recommendation system under the Department of Science and Technology Government of India.

Tech Stack Used: Tensorflow, Python, Scikit-Learn, Numpy, Pandas

SOFTWARE ENGINEERING INTERN Addverb Technologies, Noida June 2019 – Aug 2019

Optimized and enhanced the performance website’s code to optimize the response time of the web app. Handled bugs and added various new features in the UI.

Enhanced a web application's security by fetching and decrypting the data received from the API using the RSA’s algorithm.

Tech Stack: JavaScript, Angular 7.0, HTML, CSS

POSITION OF RESPONSIBILITY

ASSOCIATE DEVELOPER DTU Times (DTU's official Newsletter and Magazine), Delhi

Involved in the development of the front-end of the DTU Times website https:/dtutimes.dtu.ac.in/

Managed and guided a talented team 2nd-year developer for further development and improvement of the official website.

Tech Stack: JavaScript, ReactJs, SCSS

PUBLICATIONS

S. Gupta, A. Gupta, Anjum, and R. Katarya, “InstaCovNet-19: A deep learning classification model for the detection of COVID- 19 patients using Chest X-ray,” Appl. Soft Comput., p. 106859, Oct. 2020, doi: 10.1016/j.asoc.2020.106859.

S. Gupta, A. Gupta, Anjum, and R. Katarya, “Recent Trends of Fake News Detection: A Review,” Springer LNEE, Dec. 2020 PERSONAL PROJECTS

IMAGE DENOISING USING DEEP LEARNING: Implemented five research papers from top-level conferences and transactions on image denoising and compared their results by training and testing on the chest x-ray dataset. The research works were implemented from scratch using Tensorflow deep learning framework in python. The highest PSNR achieved from the implementations was 30.29 dB. The rest of the results and the implementation can be seen on the GitHub repository. Project-Link

IMAGE CAPTION BOT: Image Caption Bot give captions or description of the image uploaded by the user. It uses a perfect blend of both NLP(LSTM) and Computer Vision techniques (CNN and Transfer Learning(ResNet50)) to predict the caption using Tensorflow 2.0 framework. The weights make the embedding layer of the Glove Vector (by Stanford). Project-Link

FAKE NEWS DETECTOR & GENUINE NEWS GENERATOR: Developed and Deployed a machine learning model for fake news detection using naïve bayes algorithm, worked on the front-end interface, machine learning algorithm, and also deployed the website and the API. Tech Stack: Flask, Python, ReactJS; Website Link; Video Demonstration; Project-Link

AUTOMATIC MASK DETECTION SYSTEM: This model detects if a person is wearing a mask using a real-time video feed. Useful in catching mask rule violators. Uses OpenCV's Haarcascade for detecting faces and uses a deep learning model based on the MobileNetV2 model developed on Tensorflow 2.0 to detect a person wearing a mask with 97% accuracy. Project Link

TECHNICAL SKILLS

Programming Languages (C++, JavaScript, Python)

Full-stack Web Development (MERN stack, MEAN stack (Familier), Flask (Familier))

Machine Learning (Libraries: ScikitLearn, Scipy)

Deep Learning (Frameworks: TensorFlow, Pytorch)

Data Analysis (Libraries: Numpy, Pandas, Matplotlib, Seaborn) COURSEWORK AND CERTIFICATES

TENSORFLOW DEVELOPER SPECIALIZATION (COURSERA); DATA SCIENCE MASTER COURSE (CODING BLOCKS); DATABASE MANAGEMENT ESSENTIALS (UNIVERSITY OF COLORADO); IMPROVING DEEP NEURAL NETWORKS: HYPERPARAMETER TUNING, REGULARIZATION, AND OPTIMIZATION (COURSERA); SEQUENCE MODELS (COURSERA)

Contact this candidate