Research Interests Natural Language Processing, Search Engines and Information Retrieval, Machine Learning and Information Integration.
I am pursuing my Master's in Computer Science at University of Southern California. I am very interested in data sciences. I would like to perform research in applications of natural language processing and machine learning. I have worked in various domains of Computer Science such as Mobile Technologies, Web Technologies, Data Sciences, Computer Networks and like keeping abreast with the trending topics in these areas.
Leveraged the Apache Solr index built in the above project and the D3.js data visualization technology (http://d3js.org/) to interact and visualize our search engine data. Leveraged D3.js and the geolocation information from the data to recreate maps with jobs postings. Integrated the geonames.org dataset in order to get country level classifications of jobs. Developed web-based REST access to the Solr data in a compatible way with D3.js
This project was an introduction to Oracle Spatial objects and how to query maps. We used the USC map and
saved each building as polygons in oracle. Using a java based application, the user can now draw random polygon
and query the intersecting buildings.
The map had buildings and fire hydrants across USC. One could simulate a buulding to be on fire and query the
nearest fire hydrant.
Implemented an Averaged Perceptron classifier to perform part of speech tagging and named entity recognition.
Developed scripts to train a POS tagging model (postrain.py), and to tag new text (postag.py) and performed classification using the averaged perceptron. In the Part of Speech tagger, for a given sentence, each word is tagged according to its function in sentence. For e.g., Noun word is tagged with NNP.
Developed an Named Entity Recognizer (NER) and utilized the averaged perceptron which are together capable of identifying and classifying the named entities like persons, organizations and location with BIO-encoding
Developed an approach for detecting and correcting errors involving the use of words that sound similar or the same are fairly common. For example, the word its is frequently used where it's should, and vice-versa. Other confusable pairs include: they're/their/there, you're/your and loose/lose.
Used the open-source nltk library for pos-tagging . Constructed a feature vector with previous two tags and words and next two tags and words.
Built models for the confusable word pairs using megam. Classified the correct word among the pair using megam and replaced it in the orginal text.
Developed an E-commerce Information System with Administrative and Client-side functionalities for buying and selling products
using PHP, AJAX, jQuery, JS, HTML, CSS, MySQL. Implemented collaborative filtering for recommending similar products.Implemented the MVC model using CodeIgniter. Implemented Responsive Web Design for platform independent rendering using
jQuery Mobile. Secured the website from cross-site scripting (XSS) and SQL injection.
Publication : 2012 IEEE International Conference on Computational Intelligence & Computing Research (ICCIC)
Abstract : In this paper, we discuss the advent of mobile cloud computing and its advancements. It presents a review on the concept of Mobile Cloud Computing, features, statistics and growth of cloud computing, consumption of mobile services, issues in mobile cloud computing as well as prospective solutions, future scope and research. In the future scope, we discuss that the advanced mobile cloud computing platform has not only led to an improvement in the processing of mobile applications and services but has also opened a revolutionized realm wherein mobile cloud computing can be applied to the operating systems installed in mobile devices and the operating systems as a whole can run on the cloud framework, increasing the processing speed and optimizing operations to attain efficient results.
Publication : International Journal of Engineering and Technology (IJET)
Abstract : In this paper, a real-time location tracing application has been discussed to cater the needs of the people who distinctly and constantly want to stay connected with their peers as well as it can be used in disaster management situations. It is developed on the BlackBerry platform provided by RIM(Research In Motion) which is a highly secure platform and is used to mitigate security risks. The application is integrated with Blackberry Messenger which has a large number of social networking users which is a value added feature of this application. The function of this project is to trace a number of people even in an environment buzzing with hoards of people. This application will also provide the user with features such as data-sharing, forming and managing groups, communities and other location based services.
Worked as part of the Analytics cloud at Salesforce in the Wave iOS team.
Project: I created a Jenkins job which automates the entire release coverage process and can be used across all the Salesforce instances since the release coverage process was earlier performed manually.
It increased productivity by 50%.
Designed and developed a core-release suite which covers all the top use cases ex. creating dashboards, verifying lens, dashboards visibility, deep link use cases
The release suite is executed by triggering the Jenkins job which generates an automated summary of results and helps the team to predict further actions and perform analysis on those reports.
I also wrote library methods for the Server side functionalities by leveraging Wave REST APIs which can be reused.
My other tasks were participating in the sprint reviews and contributing to daily tasks
Gained exposure to the Salesforce cloud system and deployment of apps across multiple instances
University of Southern California Graduate Research Assistant Fall 2014 - Present
Courts and the Globalization of Law Project
I am a Programmer for the Courts Globalization Law Project I am developing localization modules, country specific statistics and citations for international court documents.
Developed a Java application to provide analytics on international legal citations in the decisions of international courts
Applied NLP techniques, regex and pattern matching to extract citation references in court documents of 20 countries
ITS, FMS, University of Southern California Systems Analyst Spring 2014 - Summer 2014
Performed data migration from Appsense to Druva InSync software, file backup, file sharing, troubleshooting requests
Cybage Software Pvt. Ltd., India Software Engineer October 2012 - December 2013
Developed, debugged and tested Android mobile applications, APIs and Android-compatible Java libraries
Cybage Software Pvt. Ltd. , India Software Engineer Intern July
2011 - May 2012
Team Lead for the Bachelor's of Engineering Final Year Project. Developed a mobile based Location Tracer with Social Networking Services
....to be filled
The University of Southern California does not screen or control the content on this website and thus does not guarantee the accuracy, integrity, or quality of such content. All content on this website is provided by and is the sole responsibility of the person from which such content originated, and such content does not necessarily reflect the opinions of the University administration or the Board of Trustees