Graduate / "Every day you may make progress. Every step may be fruitful";SOP (Computer Science [3]

Statement of Purpose

"Every day you may make progress. Every step may be fruitful. Yet there will stretch out before you an ever-lengthening, ever-ascending, ever-improving path. You know you will never get to the end of the journey. But this, so far from discouraging, only adds to the joy and glory of the climb".

These words of Sir Winston Churchill sums up my feelings when I reflect on the decision to pursue graduate studies in Computer Science at your university as my next career objective. Having completed my graduation with Honors, I feel my undergraduate curriculum has given me sufficient exposure to all the various possibilities for research that are possible in this field and has left me with an insatiable thirst to get more understanding of the subject in depth and engage myself in research that has bearing on the life of a common man.

I joined IIIT-Allahabad, after being selected among the top 1% in All India Engineering Entrance Exam. The breadth of the IIIT syllabi gave me a comprehensive exposure as well as a sound conceptual understanding of the core areas of Computer Science including machine learning, robotics, data structures, automata theory, networking, databases, operating systems and core languages like C/C++, Java etc. My interest in Artificial Intelligence developed while creating an expert system for the classification of a given sample of soil entered by the user. It was based on 13 soil attributes and used 61 rules to categorize the user-input into 6 different classes of soil. The output from the inference engine determined the effectiveness of cultivation. The implications of the project and the vast areas of research that the field promised, motivated me to choose Artificial Intelligence as my field of work. I learnt important concepts such as fuzzy logic, neural networks, evolutionary computations and optimizations. I also employed techniques like K-Means Clustering, Self-Organizing Maps and Genetic algorithms on test-datasets during my lab hours that helped me build-up an interest in the field.

In line with my interest I worked as a Junior Research Associate under Dr. Ratna Sanyal, Coordinator of the Universal Digital Library (UDL) Project, IIIT - Allahabad. A word about the project, the major objective of the project is to capture and store more than a million books in the digital format and to develop the required language technologies to support the same. The initiative is taken by a group of institutions spread across different parts of the country together with the Carnegie Melon University (USA). As a part of this project, my first task was to identify the sense of the given polysemous word in the user text corpus. During an exhaustive study of the approaches that have been taken to solve this problem, I came across Yarowsky Algorithm, a semi-supervised, bootstrapping algorithm, which claims to have a high precision (above 90%) as compared to the other existing methods, mainly for an open-text set of data. Based on the results of the project, I came to the conclusion that the algorithm is of a great practical value mainly in cases where accuracy from coarse grained disambiguation techniques is unsatisfactory. The research involved exposed me to important NLP concepts such as Stemming, Stop words removal, Part-of-Speech Tagging, Hidden Markov Models and N-gram models.

My next project was based on Information Retrieval. The task was to provide a multilingual indexing service for an aggregative digital library to be used in a distributed environment. This project was carried out as a part the Technology Development for Indian Languages (TDIL) Programme initiated by the Department of Electronics & Information Technology (DeitY), Ministry of Communication & Information Technology (MC&IT), Govt. of India. Inverted indexing technique was used to store the unique terms present in the given text document, along with a list of documents having those terms. While a word-level inverted index could have been used as it provides more functionality by giving the positional information of the key terms, a record-level inverted index was used as it involved lesser space and time overheads. Additionally my approach also supported dynamic indexing which is suitable for environments where the documents are frequently changing.

During my student internship at Motorola, I created a domain specific language for a video controller device that provided secured access control for digital video distribution. I used JAVA based ANTLR libraries for creating the parser as it was much easier to understand and debug than the traditional YACC/Bison parsers, that I had worked on during my lab assignments in the "Compiler Design" course. I also created specification for its syntax and provided hooks for integration with backend-plugins. Knowledge of Data Structures was used for storing and processing of the data during the whole process.

Besides these projects, I have also worked in the field of computer graphics. This involved creating an interactive visualization of large medical volumetric datasets using multi-resolution bricking. A user could interactively explore the dataset by specifying a region of interest (ROI), which could further generate a much more accurate data representation inside it by use of nested ROIs. VTK, the visualization toolkit was used to perform the image rendering and a shader language was used to speed up the rendering process. For creating sub-samples of the large-sized medical input, so that they could be visualized by the rendering technique, STXXL (C++ Standard Template Library for Extra Large Data Sets) libraries were used. I also implemented an OpenGl program where I used the C based glut libraries to perform animation of a 2D robot on different types of terrains.

I have been working as a Business Technology Analyst in Deloitte for the past two years. I am part of the Digital Content Management Team which provides business, technology & creative expertise to deliver digital business solutions for some of the world's leading brands. My role is to develop modules that provide an easy-to-use application that would enable the business users to author and publish content across channels, while maintaining their corporate standards. For this Adobe CQ Web Content Management (WCM) platform is being used which provides a productive, easy-to-learn authoring environment with support for in-place editing; drag-and-drop page composition from a rich library of web components; and intuitive controls for search-engine optimization, scheduled delivery, and landing page optimization including real-time and multi-variant tests. This platform is based on JAVA as the programming language. Besides being involved in the development related activities in the firm, I have also been part of various firm-related activities aimed at promoting social welfare as well as developing good organization skills.

I believe that a good graduate program is an essential step for realizing my professional ambition of becoming a competent researcher. I also believe that I possess the motivation, intellectual ability and preparation to do justice to a demanding graduate program. My excellent undergraduate record and current work experience are a testimony to that. I have spoken to my professors, seniors and colleagues about my choice of University, and have applied to the XYZ because it is reputable for its impressive placement opportunities, room for technical growth and computing resources. The department web site revealed a very extensive research in the area of Artificial Intelligence. It has got outstanding research infrastructure and learning computer science under your esteemed and award winning faculty would indeed be an honor for me.

I appreciate the pedagogy which does not harp on conventional classroom teaching methods but rather compels the students to think out of the box. Your computer science curriculum caters to my requirement of studying subjects of my interest, from artificial intelligence, networking to software systems which will help upgrade my knowledge base and aid in my complete understanding of computer science. I am hopeful that my confidence, aptitude for learning, strong determination, resilience, and positive approach will facilitate my growth as a scholar.

I bring with me a strong grasp of fundamentals in Computer science, an aptitude for research and team work, a zest for challenges and an enthusiastic desire to learn all I can. I would like to take with me, in addition to knowledge of Computer System, a network of strong and lasting relationships with my teachers and fellow-students. I hope that my background and qualifications are found suitable for MS in Computer Science at XYZ University.