Monik Bansal Email & Phone Number

United States

Delaware, United States

Delaware, United States

United States

• Researching and developing algorithms to perform unsupervised analysis on twitter data produced during natural disasters.• Fetched 0.5M tweets from 5 countries during COVID‑19 and performed LDA Mallet Topic Modelling on the dataset.• Created Word2Vec model and obtained a single vector for each tweet. Implied K‑Means to group similar tweets.• Generated time series graphs to analyse the results and draw insights for research.

New Jersey, United States

- Researched and developed machine learning models using agile software development methodology under the mentorship of experienced ML practitioners.- Worked with clients from various industries such as FinTech, Healthcare and Fashion to build end to end machine learning solutions from data collection, EDA to prototyping and deployment.- Worked on Fraud detection and recommender systems using PyTorch and Fast.ai libraries.- Improved the production-level multi-modal visual-textual search engine and evaluation metrics for one of the reputable brands.-Used Deep Learning model Resnet‑50 Multilabel Classification to identify the manipulations in a document and achieved 77% accuracy.-Worked in collaboration with Platform.AI to detect varying levels of doneness in pizzas with an accuracy of 91% on the client’s curated test set. Employed various data augmentation techniques like Mask R‑CNN and Facebook’s DETR to build a synonymous training set.

Master'S Degree, Data Science, 3.65/4.00

- Learned and implemented projects involving machine learning and deep learning models and framework such as linear regression, SVM.

Bachelor'S Degree, Computer Science And Engineering

Data Structures, Data Mining, Machine Learning, Database Management Systems, Probability