Sandesh Swamy

Graduate Research Associate
Computer Science and Engineering
The Ohio State University
Email: swamy(dot)14@osu(dot)edu
Follow @padfoot_29 on Twitter
Here is how my name is pronounced.

I am a Master's student at The Ohio State University in the Department of Computer Science. I am advised by Dr. Alan Ritter and Dr. Marie-Catherine de Marneffe. My research involves Natural Language Processing and Social Media data. I was previously the Teaching Associate for Introduction to Computer Programming in C++ for Engineers and Scientists.

My areas of interest include Natural Language Processing, Data Mining, Information Extraction from Social Media, Artificial Intelligence, Algorithms and Data Structures. I have significant programming experience in Java, C and Python (code samples and HackerRank profile can be found in the Navigation bar).

I am also fascinated by the amount of information generated and gathered on Social Media. It is wonderful to see that news and alerts can get propagated fairly quickly with the use of Social Media tools like Twitter. With great power also comes great responsibility. Analyzing of Twitter data is a keen interest area of mine since it helps get a better understanding of the kind of things that people are really interested in and want to talk about.

Skills and Expertise
  • Programming Languages : Java, Python, C, C++, MySQL, JavaScript(beginner), R(beginner), HTML(beginner), PHP(beginner), Android Programming(beginner)
  • Tools and IDEs: Eclipse, Visual Studio, R-Studio, Android Studio, Sublime, PyCharm, Jupyter, Github, Octave
  • Although I do not claim to know all the Linux commands off the top of my head, I do love working on Linux and the charm of the plain old Terminal cannot be beaten by any IDE
  • My favorite programming languages off late have been Python and Java. Super powerful.
Academic Projects
  • At The Ohio State University
    • As part of the Data Mining coursework in my first semester at OSU, I worked on kNN classifiers, k-means clustering and Decision Tree classifiers. The coding was done in Python and R and used packages like numpy, scipy and matplotlib. It can be accessed here. (Python, matplotlib, numpy, scipy, Weka, R)
    • As part of the Computational Linguistics coursework in my first semester I had the opportunity to learn and work on CKY parser implementation, Naive Bayes text classification (Sentiment Analysis), Zipf's law verification for text documents, entropy and its significance and a host of other topics. Some of the implementations can be accessed here. (Python)
    • As part of the Speech and Natural Language Processing coursework, I experimented with Sentiment Analysis of IMDb movie reviews using Naive Bayes Classifier and Averaged Perceptron. I also had the opportunity to implement the Viterbi algorithm to perform POS tagging on large datasets which included Twitter and IRC data.
    • Carried out a project (group of 4) titled "Feature based classifier for differentiating poetry and prose". I was responsible for the baseline classifier which achieved an F1 score of ~85%. (Python, numpy, scipy, CRFSuite)
    • Implemented the EM algorithm on a 1-dimensional dataset using R.
    • Carried out a project (group of 3) titled "Public Perception to currently showing movies". I was responsible for the entire design and code. Used TMDb API for fetching movie data and cutting down the number of popular movies based on the popularity index. Then used the Tweepy library to gather tweets with a combination of the words which would be used as a hashtag for the movies. Further, TextBlob API and Naive Bayes Classifier were used to perform Sentiment Analysis. (Python, TextBlob, TMDb API, Tweepy library)
  • At R.V. College of Engineering
    • Carried out a project (group of 4) titled "Efficient Information Retrieval using Crawling, Indexing and Ranking". I was responsible for the crawler module which was used to gather links related to a search query. (Java, JSP)