I'm Madhumitha, currently a senior pursuing Information Technology. I come from Visakhapatnam, Andhra Pradesh. I am a developer and machine-learning enthusiast, interested in making ML accessible to everyone.
I play Chess and Basketball and love organizing and creating. My hobbies include reading books and writing poems and short stories. I am passionate about music, rarely seen without headphones on.
I am interested in Web Development and Machine Learning.
Corporate Exposure
Predicted Execution times of Spark SQL queries to optimize job scheduling
Implemented a pipeline to collect feature metrics of a query using Spark REST API
Implemented an XGB model to predict the execution time of the query with an accuracy of 90.2%
May 2020 - July 2020
Built a packet capturing application for Windows
It stores all the packets that are transmitted on the chosen connection.
A VC++ appication that uses WinPCap
May 2019 - June 2019
Papers published so far
With the increase in the number of open-source and commercial code in the market, copyright and license infringement cases are on the rise. The lack of availability of source code makes identification a difficult task as existing techniques heavily rely on the source code. We propose two similarity measurement methods using the instruction log and the call-trace of each executable using Intel Pin tool.
Accepted in the ISSRE Conference, 2020
Proposing an auto-encoder approach enhanced by K-Means Clustering preprocessing to improve compression rates of data with minimal loss. Using clustering resulted in a net gain of 1% over the standard encoder architecture over three MNIST datasets. The compression ratio achieved was 24.6:1.
Accepted in the FICTA Conference, 2020
Comparative study of deep learning software frameworks such as Torch and TenserFlow (with Keras API)
Benchmarks the running performance of the frameworks with three types of neural networks
Accepted in the ICOEI Conference, 2019
• Implemented a novel CNN model that can classify an X-Ray as COVID -19 positive or negative with 99.2% accuracy and 96.00% sensitivity
• Implemented a novel CNN model that can classify an X-Ray into healthy, COVID-19 positive or Pneumonia positive with 86.13%accuracy and 88.00% sensitivity for COVID-19
• Currently working on extracting more information from the X-Rays such as severity and other relevant information
• Built a Django App from women to monitor their weight, sleep cycles, heart-rate, calorie intake and other health cycles.
• The app reduces number of clicks to start a workout, hence making it easier to start exercising.
• The app was built at a Microsoft hackathon, and received praise for the thought put behind the design of the web app.
• Implemented an Solidity-based decentralized app for small-time digital content creators to copyright images, audio and text,thus preventing intellectual property theft of tweets, SoundCloud uploads, etc.
• Implementation streamlines the copyright defense process, thus eliminating legal fees and time costs
• Worked on the Django app that takes in a multispectral mutli-temporal dataset to extract crop parameters
• Python modules such as rasterio and OpenCV were used to separate the spectral data while matplotlib was used to generategraphs
• The crop season (kharif, rabi, zaid) was identified along with other useful crop details
A few interesting projects over the years
Bachelor of Technology in Information Technology. Will graduate in 2021. Currently in my final semester with a CGPA of 9.03
CBSE Class X, passed out with a CGPA of 10/10
Board of Intermediate Education, Andhra Pradesh, passed out with a percentage of 96.4%
Classes I to V
Where it all began
What I do when I'm not coding
Responsibilites include leading a team of 70 members spread across six teams and providing the best environment for the 700 participants for the event
Responsibilites include leading a team of 170 members spread across seven special interest groups and conducting recruitment drives, technical events, discussions and networking opportunities
Responsibilites include organizing tournaments and team-managing
This article talks about one of the possible (probably the hardest) ways of earning one million dollars. The Clay Mathematics Institute offers a prize of $1,000,000 for the first correct solution of any one of its seven Millennium Prize Problems. One of its seven problems, and probably the easiest to explain, is the P vs. NP problem.
Read some of my thoughts