I started my journey of learning from Parvatipuram, AP which continues even after attaining an M.Tech in CSE from IIT Bombay. My M.Tech project involved Automatic Ontology Generation, which increased my interest on Information Extraction and Machine Learning techniques for unstructured data. More about my B.tech and M.Tech projects can be found here. Currently, I am deeply involved with "Cognitive Solutions and Services" group, at IBM India Research Lab. Prior to joining IBM I was engagged with Strand Life Sciences,in developing automatic identification of relations between various biological molecules and processes. More about my work can be found here.

Academic Background
ExaminationUniversityYearCPI/%
Post GraduationIIT Bombay20109.12
GraduationBiju Patnaik University of Tech.,Orissa200578.88
Intermediate/+2Board of Intermediate Education, A.P.200190.4
MatriculationBoard of Secondary Education, A.P.199979.33

Research Interests
  • Natural Language Processing
  • Pattern Recognition
  • Machine Learning and Data Mining

Projects Of Own Interest

Speech to text converter - Extension for UNIX:
  • Trained Sphinx4 so that it can recognize Indian speaker voices. And used it to convert vocal commands of user to execute in any unix based systems.

Suggesting preferable seat using Regression:
  • Predicting suitable seat for a customer while booking for a movie in a movie theatre using online booking system. Solved the program using regression.

Supervised techniques to assist product selection:
  • Predicting whether a movie or a product that is going to be released in the market will be a hit or flop. Used automatic ontology construction to find the relavent information required to deide. Used sentiment analysis on relavent data regarding the movie/product and then used SVM to tell whether it is a thumps up or thumps down.

Projects

Building domain ontologies from lecture notes:
(M.Tech thesis, under the guidance of Prof. Sridhar Iyer)

slides
  • Developed an automatic tool which will develop the ontology from the given lecture notes or text files. The tool also can give the possible pre-requisites and follow-ups for a given learning module and it will provide a concept map to the user so that he/she can have a conceptual view of the domain.

Study of different tools, languages and methodologies for developing ontologies:
(M.Tech seminar, under the guidance of Prof. Sridhar Iyer)

slides
  • A comparative study of different tools and languages available for developing domain ontologies.

Projects

Building domain ontologies from lecture notes:
(M.Tech thesis, under the guidance of Prof. Sridhar Iyer)

slides
  • Studied different repositories and different searching techniques of the repositories.
  • Studied different NLP tools and Mining Techniques to find out the relations between the keywords which are extracted from the corpus text.
  • Developed an automatic tool which will develop the ontology from the given lecture notes or text files. The tool also can give the possible pre-requisites and follow-ups for a given learning module and it will provide a concept map to the user so that he/she can have a conceptual view of the domain.

Study of different tools, languages and methodologies for developing ontologies:
(M.Tech seminar, under the guidance of Prof. Sridhar Iyer)

slides
  • A comparative study of different tools and languages available for developing domain ontologies.
  • Reviewed different methodologies available for developing ontologies.
  • Developed sample ontology using Protg tool in OWL (Web Ontological Language).

Implementation of IWSD:
(under the guidance of Prof. Pushpak Bhattacharyya)


Construction of pronunciation dictionary from existing dictionary:
(under the guidance of Prof. Pushpak Bhattacharyya)


  • The goal here is to "learn" pronunciation for character sequences from existing dictionary (cmusphinx dictionary) and generate pronunciation for new words which were not present in the dictionary. Used giza++ to obtain the phase table and Moses tool to produce the pronunciation.

Prevention of cross site scripting:
(under the guidance of Prof. Bernard Menezes)


  • Implemented 'Blueprint' a XSS prevention technique from the server side. It will enable a web application which effectively takes control of parsing decisions and automatically creates a structural representation of untrusted web content that is free of XSS attacks.

Tesxt to speech converter, Firefox ADD-ON:
(under the guidance of Prof. G. Sivakumar)


  • Developed a Mozilla Firefox Add-on which will play the selected text area on the firefox web browser using eSpeak Speech synthesizer for English.

Simulation of Bit Torrent:
(under the guidance of Prof. Kameswari chebrolu)


  • Implemented the functionalities of BitTorrent, a peer-to-peer communications protocol for file sharing, using Socket programming in C.

Simulation of snoop TCP:
(under the guidance of Prof. Anirudha Sahoo)


  • Comparative study of Snoop TCP and classical TCP Reno through simulation results. Both Snoop TCP and TCP Reno are implemented using NS2.

Product news release:
(under the guidance of Prof. G. Sivakumar)


  • A web based application with the mission of delivering timely and new industrial product information covering the whole range of products.

Online issue management:
(B.Tech Project)


  • A web based project to facilitate the employees of company to communicate with each other through Net Conferencing. The project is developed using ASP.Net 2005 and Oracle 8i.

Structure of hydro carbons:
  • Displays the graphical representation of Chemical Bonding between the carbon and hydrogen atom in a carbon compound for a given number of Carbon and hydrogen atoms using C Graphics.

Seminars

Study of different tools, languages and methodologies for developing ontologies:
(M.Tech seminar, under the guidance of Prof. Sridhar Iyer)


  • A comparative study of different tools and languages available for developing domain ontologies.
  • Reviewed different methodologies available for developing ontologies.

Dependency parsing:
(Seminar, under the guidance of Prof. Pushpak Bhattacharyya)


  • A comparative study of different types of parsing techniques that were avilable for dependency parsing.
  • Study of different evaluation techniques that were avilable for dependency parsing.

Seminars

Study of different tools, languages and methodologies for developing ontologies:
(M.Tech seminar, under the guidance of Prof. Sridhar Iyer)


  • A comparative study of different tools and languages available for developing domain ontologies.
  • Reviewed different methodologies available for developing ontologies.
  • Developed sample ontology using Protg tool in OWL (Web Ontological Language).

Dependency parsing:
(Seminar, under the guidance of Prof. Pushpak Bhattacharyya)


  • A comparative study of different types of parsing techniques that were avilable for dependency parsing.
  • Study of different evaluation techniques that were avilable for dependency parsing.
  • Pros and cons of dependency parsing.

Automated building of domain ontologies from lecture notes in courseware:
(Paper presentation, along with Prof. Sridhar Iyer)

 IEEE Conference on Technology for Education (T4E), Chennai, India, July 2011.
  • Study of automatic ontology construction.
  • And demo of an automatic ontology tool that constructs ontology from courseware repository which will hel student in finding the pre-requisites and follow-ups for a specified topic.

Experience

IBM India Research Lab, Delhi, India :
(Oct, 14 to Till date)

 Currently working as a Software Engineer in Cognitive Solutions and Services group. Which deals with Exploratory data analysis

Strand Life Sciences Pvt Limited, Bangalore, India :
(July, 11 to Oct, 14)

 Worked as a Specialist software engineer at Strand life sciences. My work deals with automatic identification of relations between various biological molecules and processes.

Sutherland Global Services India Pvt Limited, Chennai, India :
(July, 05 to Nov, 07)

 Worked as a Subject Matter Expert from Jul 2005 to Nov 2007 for US process, HP Notebooks.

Experience

IBM India Research Lab, Delhi, India :
(Oct, 14 to Till date)

 Currently working as a Software Engineer in Cognitive Solutions and Services group. Which deals with Exploratory data analysis.i

Cognitive AMS: A Watson based cognitive agent assist tool that facilitates users with faster resolution of application related L1/L2 tickets. My major contribution to this project includes enabling search based on knowledge graph, and in development of user interface. For construction of knowledge graph from a given dictionary of error messages, used natural language dependency parsing to extract entities, actions, errors to construct error graph from textual description of errors. Used this error graph to drive conversations to arrive at a particular error message from the user query.

Java skill assessment: Given a developers software projects, assesses programming language proficiency in terms of features used, Code quality, data-structures used, Third party components used. Further, the system collects benchmark data from publicly available software projects (e.g. GitHub, JazzHub, GoogleCode etc). Based on benchmark data, provides relative ranking for a developer when compared to the community. And the system also provides recommendations on how to improve language proficiency.

Exploratory data analysis: This project consists of two aspects, “Problem Tickets Analysis” and “Data Exploration”. “Problem tickets analysis” deals with analyzing domain independent, unstructured information present in the tickets logged for issues ranging bugs to clarifications. This data is parsed and hosted in web interface for the statisticians to get insights out of it. On the contrary “Data exploration” supports the need of endusers, by allowing upload of data sets to leverage the web based slicing and dicing features without any text analysis applied.



Strand Life Sciences Pvt Limited, Bangalore, India :
(July, 11 to Till date)

currently I am working as a Specialist software engineer at Strand life sciences. I am currently dealing with Interaction Database, Which is a database of relations between various biological molecules and processes. The molecules and processes are considered as "Entities" and their biological interactions as "Relations". We consider well know organism's like Human, Mouse, Rat, Rice, and Yeast etc., for our Interaction Database. For the identification of Relation we use Natural Language Processing (NLP) algorithm that runs on published Medline abstracts. The algorithm is based on a 'deep parsing' and it operates on a sentence-by-sentence manner to extract the relations with in a sentence. Broadly there are two areas

  • Entity Tagging: Entities for our analysis are the properties of proteins, process, family, function, small molecules, complex, and enzymes in the biology world. The properties were identified from different sources like ncbi, nlm and gene ontology repositories. The properties were tokenized in ourder to identify all forms of the same term. One more feature incorporated in tagging is, identifying local abbrevations.
  • Relation Identification:After identifying the entities in a given sentence the next step is to find the relation between the entities. Used context free grammer to identify the syntax of the sentence (syntax tree). From syntactic tree we identified the semantic tree using a predefined dictionary. From the semantic tree we identified the relations amont the entities. In our case we have fixed number of relations like regulation, binding, member, modification etc.


Sutherland Global Services India Pvt Limited, Chennai, India :
(July, 05 to Nov, 07)

 Worked as a Subject Matter Expert from Jul 2005 to Nov 2007 for US process, HP Notebooks.
  • Resolution Expert :
    (from July, 05 to May, 2006)
    • Handling calls for US HP Notebook customers, providing resolution for customer queries on both Hardware and Software related issues.
    • Performing the installations, configuring and networking of systems, devices, operating systems and applications throughout the network.
    • Troubleshooting and resolving operational issues of LAN/WAN, Internet/intranet services, remote accessing
  • Resolution Specialist :
    (from May, 2006 to Nov, 2007)
    • Handling the team by providing support to team members, by providing resolution for unresolved and critical issues to the customer where the call back is committed.
    • Worked closely with Software team and customer to determine functional requirements and developments for customized solutions
    • Developed excellent rapport and built relationships with the team members as well as the customers through attention to details in defining need and providing effective support and resolution.

Other Roles & Responsibilities
  • Worked as a member in the organizing team of T4E (International Conference on Technology for Education) in July, 2010.
  • Worked as a Consultant Teaching Assistant for Undergraduate course 'Computer Programming and Utilization' under Prof. Deepak B. Phatak, responsible for conducting tutorial sessions, helping the students with their doubts, grading quizzes and semester examinations.
  • Teaching assistant for Undergraduate course Data Structures and Communication Networks under Prof. Sridhar Iyer, responsible for helping students during tutorial sessions, grading quizzes and semester examinations.

Achievements
  • Secured 99.73 percentile in GATE-2009.
  • Qualified NCC A Certificate Examination under the authority of Ministry of Defence, Govt. of India.
  • Was awarded as Best Performer for maintaining excellent CSAT and sales for the month of Aug, 2007.
  • Represented school at District level in “Dist. Educational Science Fare” from 1997 to 1999.
  • Actively participated in Janmabhumi Scheme, a programme involving people in the implementation of welfare measures. And received best volunteer award from the Collector for the year 1997.
  • Worked as Facilitator for Akshara Sankranti, Vizianagaram 2000, A district level Literacy program, from 2-10-2000 to 31-3-2001.

Categories :

Contact Me :

IBM India Research Laboratory
Plot No.4, Phase - II,
Block - C, ISID Campus
Institutional Area, Vasant Kunj,
New Delhi-110070