Madhura Das
Graduate Student at Carnegie Mellon University
Specializing in Natural Language Processing and Machine Learning
I am a Master's student at Carnegie Mellon University. I have a keen interest on solving problems with the help of Machine Learning and Natural Language Processing. My previous experience has been on Natural Language Generation and Time Series Analysis. I also have a strong background on web development and UNIX system admin tasks
| Degree | Institution | Years |
|---|---|---|
| Master of Science in Intelligent Information Systems | Carnegie Mellon University, Pittsburgh | 2017-Current |
| Bachelor of Engineering | PES Institute of Technology, Bangalore | 2012-2016 |
| Secondary Education | National Public School, Bangalore | 2010-2012 |
Advances in Computing and Communication Engineering (ICACCE), 2015 International Conference, Dehradun, India
Publisher: IEEE
Parallelization of Searching and Mining Time Series Data using Dynamic Time Warping
Published in Advances in Computing, Communications and Informatics (ICACCI), 2015 International Conference, Kochi, India
Publisher: IEEE
Automated Content Suggestion for Document Writing
Published in International Conference on Computational Intelligence and Computing Research (ICCIC), 2016, Chennai, India.
Publisher: IEEE
Automated CAPTCHA Generation from Annotated Images using Encoder Decoder Architecture
Published in International Conference on Information Technology (ICIT), 2016, Bhubaneshwar, India
Publisher: IEEE
Scores and players of 100 cricket matches were scraped from the web using Python. A probability based algorithm was developed to predict the winner of a simulated match using Apache Mahout clustering and Apache Spark MapReduce algorithms. The results had a 70% accuracy.
Summarization and Text Analysis
Cricket commentary of a match was scraped from the web using Python. A score board for the match was created by analysing the commentary text. A maxEnt model was developed for event detection from the ball by ball commentary. An executive summary of the match was generated using natural language generation techniques.
Green Cloud Swift
A solution was provided to improve the energy consumption of OpenStack Swift which is the data storage module belonging to OpenStack. The solution developed was to dynamically bring up nodes periodically to maintain the data consistency across the cloud and on other time have only 1 data node service the cluster.
Derby Management System (Team Lead for team of 10 members)
A complete Derby Management System was developed with the local Derby as a customer base. The backend was implemented by using MySQL as a database with server side scripting by PHP. The user interface was created using HTML and CSS. A betting solution was proposed for members of the derby through a web interface and Android application.
- "Prof. MR Doreswamy Distinction Award" for scoring a GPA above 9.0 in all eight semesters during my undergraduate studies
- 3rd rank in Computer Science by scoring 96 out of 100 in 12th board exams
Organized monthly charity drives in my local community whose proceedings went towards the education of underprivileged students.
Google Online Marketing Challenge
Managed a Google ad campaign for a non-profit organization (India Literacy Project) using Google AdWords to bring in more awareness about the organization
Decorated the college campus during the annual fest of Aatmatrisha within a budget. We had the responsibility of organizing creative events and assist in the judging process
Ayana Hackathon, Core team member
Provided suggestions and assisted the participants during the hackathon with their ideas. Also, helped with logistics and infrastructure setup for the hackathon.
Child Model
Featured in over 40 regional advertisements in print and video as a child. One of my advertisements.
English(Fluent), Hindi(Fluent), Bengali(Fluent)
Hobbies
Classical Indian Dance (Bharatnatyam, Kathak), Painting, Reading, Travel
