Shubhangi Agarwal

I am a research scholar in Department of Computer Science, IIT Kanpur. I have worked in Graph Mining, focusing on Approximate Subgraph Querying in both deterministic and probabilistic settings, which falls under the broad umbrella of Graph Mining. I was introduced to Graph Mining as a sub-field of Data Mining with applications in medical domain, where I would like to make a positive impact. I have also explored Graph Neural Networks, popularly known as GNNs for node and edge level prediction tasks. I am interested in applying them to various application domains, like, bioinformatics, knowledge graphs, infromation retrieval and fraud detection.

After coding in four different languages while working on various problems and projects, I realised I enjoy working in C++, the introductory programming language for me, as well as Python over C and Java. While coding in Python I have touched many libraries, PyTorch, Tensorflow, Scikit-learn, NumPy and Pandas. I also have some experience with the language R from my internship days, and was introduced to Weka for Machine Learning during my PhD course-work. During Bachelors amongst many other things (Data Structures and Algorithms, Computer Organization, Operating Systems etc. etc.), I remember learning PL/SQL, the procedural language extension for SQL, and hoping to effectively use triggers in a deployable project. Like every other programmer, I have some experience in JavaScript, HTML, PHP and Docker; and like most of the PhDs around the world I have used LaTeX enough to have written my thesis in it; is much smoother experience than most would believe, at least most of the time ;) And while I am mentioning all this, I can’t not mention this; for the past several years I have become so used to the command line programming and various parts of shell scripting, so much so that I cannot imagine myself using a Windows OS anymore!