Name: Pooja Srinivasan
Type: User
Company: Columbia University
Bio: MS Data Science @Columbia University, New York (Dec 2023) | ECE @SSNCE, India (2018-2022) | Data Science Intern @Blocp | GHC23 Scholar
Location: New York, NY, US
Blog: https://www.linkedin.com/in/poojasrinivasan11/
Pooja Srinivasan's Projects
Studied rapid growth of the Asian American Community using R and ggplot; Used D3 to build an interactive graphic, and implemented data analysis to obtain trends & patterns between social & health issues, well-being, challenges, and concerns
The objective of this project is to analyse and select the best locations in the city of Kuala Lumpur, Malaysia to open a new shopping mall. Using data science methodology and machine learning techniques like clustering, this project aims to provide solutions to answer the business development question.
Developed recommendation pipelines leveraging content-based and collaborative filtering to present top n customer recommendations from user items and customer purchase histories. Alternatively, image similarity recommendations were generated using k means clustering and Neural Networks (NNs) from product images.
The Leek group guide to data sharing
Built machine learning and time series forecasting model with extensive pre-processing & feature engineering. Leveraged Decision Tree to predict radio link failure given weather forecast conditions around radio link station for any of next 5 days with a highly imbalanced dataset.
Plotting Assignment 1 for Exploratory Data Analysis
The most predominant non-verbal communication method used to understand the mentality and the emotion of a human being primarily revolves around facial expressions, gestures, and body language . As a result, built a Python-based hybrid intelligent structure for facial expression detection, employing 7,000 images and face landmarks.
Trained GNNs to classify COVID-19 severity level by modeling states and borders in India as a graph. Combined semi-supervised learning with pre-pandemic census information such as foreign visitor count and health index of only 6 states to achieve 82% accuracy, a 15% improvement over non-graph-based models.
Built an image classification machine learning model using KNN.
To improve the adaptability of Large Language Models (LLMs) by examining and optimizing the storage paradigm within autoregressive transformer models. The emphasis is on pinpointing and editing the locations where factual associations are stored, ensuring that the models retain current and relevant information without requiring extensive retraining
Repository for Programming Assignment 2 for R Programming on Coursera
Window functions play a crucial role in Python for data science tasks, and similarly, R also offers a suite of window functions that are highly significant in the realm of data science. The accompanying PDF provides a comprehensive list of R window functions, serving as a valuable resource for professionals in the field.
Peer Assessment 1 for Reproducible Research
Project containig related material for my TensorFlow articles