Data Flow
NLP Pipeline
Data Lake
Data Cleaning
Database
Text Analysis
DBMS
Edge Prediction
Data Table
projects in
Natural Language Processing
Sriram
Information Extraction and Prediction for Drug Repurposing (2021)
Drug Discovery and Development has always been a field where efficiency is a priority, given that new solutions could serve to greatly benefit many communities. However, the different rounds of testing and research require that lots of time and resources be invested in every solution a Pharma company is exploring. This project seeks to predict which solutions deserve the most investment by performing Information Extraction on legacy data to establish past trends. Then, the data extracted from legacy research can either be compared with trends in patent records and scientific literature to form Time Series-based predictions, or inputted into Network Graph-based analysis to output new likely relationships.
Text Analysis
Building an NER (Named Entity Recognition) model to obtain "drug", "target", "disease", and
"species" entities from preprocessed text, achieving a F1 score of 98%. Extracting entities from
legacy data.
Skills : Bio-Entity Recognition
projects in
Computer Vision
Optical Character Recognition "in the wild" (2021)
Developing text recognition functionality to classify complex instances in natural scenes for a Visual Prosthesis device. See an example result here.
Skills : Deep Learning, TensorFlow, Keras, Text Detection, Text Classification
Facial Recognition and Filters (2020)
Implementing & comparing four Facial Recognition techniques: Facial Landmarks, Eigenfaces, Fisherfaces (similar to Eigenfaces), and Deep Learning using the VGG-Face Model. Read full paper here.
Skills : TensorFlow, Keras, scikit-learn, opencv-python
projects in
Systems & Data
COVID-19 Vaccine Tracker (2021)
A data visualization for COVID-19 vaccination rates in the US, updated daily, showing the rate of and number of people vaccinated in each state, made with US Census and CDC data. Used Python and GeoPandas for initial analysis, and GeoJSONs and d3.js for web application.
Written in JavaScript.
Skills : Pandas & GeoPandas, Geospatial Data Visualization, d3.js
Climate & Crop (2021)
Evaluated the relationship between climate change and national crop yield through regression analysis, sourced from NASA and PANGEA, and visualized as a Gapminder visualization.
Written in Python, SQL, JavaScript.
Skills : Data mining & wrangling, ML modeling, Statistic analysis, d3.js
Shell (2020)
Implemented a shell interface system, controlling system operations. Learned how to manage systems, concurrent procedures, and batch processes.
Written in C.
Skills : Signal Processing, Handling Multiple Jobs, I/O Redirects
Database (2020)
Built a hierarchical modeled database and implemented both client and server interfaces. Built server to handle multiple client connections while maintaining data integrity and thread-safety.
Written in C.
Skills : Networking, Multithreading, Thread Safety, Signal Handling
SINDURA SRIRAM
சிந்துரா ஸ்ரீராம் · தமிழ்
司愛悅 · 中文， 簡繁體字
Hi, I'm Sindura, a software developer based in the SF Bay Area. I'm a 2021 graduate of Brown University, where I majored in Applied Math and Computer Science.
At Brown, I have explored Computer Science as a tool for communication and analysis, beginning with my work in writing, editing, and web design -- see theindy.org, Brown's weekly art and culture newspaper. For my undergraduate capstone project in Data Science, I have tried to measure the impact of climate change on US agriculture through regression analysis of temperature change and crop yield.
I'm interested in building new solutions to solve problems old and new. Reach out to me and I'd be happy to chat with you!
