#433, Shoreham Street,
Sheffield, South Yorkshire,
United Kingdom – S2 4FB
Mobile : (+44) 7405395809
I am a computer science engineer, currently pursuing MSc in computer science with speech and language processing from the University of Sheffield. I find problem solving quite natural and I find myself creating a method on the computer to complete any given task more efficiently and effectively. I am enthusiastic, knowledge-hungry learner, eager to meet challenges and quickly assimilate new concepts. I am a highly motivated self-starter who takes initiative without supervision. I have more than 30 months of work experience in the field of Intellectual property, in particular patents. My aspiration is to grow with the organization where I can continue to learn, take on additional responsibilities, and contribute as much of value as I can.
IT SKILLS:
PROGRAMMING LANGUAGES: C, C++, Java, Groovy, PHP for scripting, Python 7.2 (for Text processing & Natural Language Processing), Visual Basic (6.0), Pure Data and MATLAB for Speech processing
VERSION CONTROL: Bazaar, Mercurial
OTHER LANGUAGES: CSS, HTML, XML
LIBRARIES & FRAMEWORKS: MVC framework, Grails framework (includes Spring & Hibernate)
CONCEPTS: Natural Language Processing, Machine Learning Techniques (such as building GMM, HMM, Support Vector Machines, etc.,), Object-Oriented Concepts and UML (Unified Modeling Language)
TOOLS: LUCENE, Stanford NER, LATEX, Bibtex, Jabref, Netbeans 6.9.1, Eclipse, Komodo edit 7, Geany, Gedit, Microsoft Office (95 through 07)
PLATFORM: Windows 98 through 7 including Windows Server editions, Ubuntu 10.10 through 11.04, Mac OS, Linux
EDUCATIONAL QUALIFICATIONS
· MSc in Computer science with speech and language processing (Expected 2011)
Expecting to pass out in first class with distinction from the 'University of Sheffield', Sheffield, UK
· Bachelor of Engineering (B.E) in Computer science and Engineering (2003-06)
Passed with first class (75%) from Arunai Engineering college (Anna University), Tiruvannamalai, Tamil Nadu, India.
· Diploma in Computer Science and Engineering (DCSE) (2000-03)
Passed with first class (72%) from Government Polytechnic college (Directorate of Technical Education), Krishnagiri, Tamil Nadu, India.
· Matriculation (10th Standard) (1999-2000)
Passed with first class (79%) from Sri Vijay Vidhyalaya Matriculation Higher Sec. School (Tamil Nadu Government Examination Board), Dharmapuri, Tamil Nadu, India.
EXPERIENCE
Dissertation: Knowledge Base Population task for participation at TAC (Text Analysis Conference)
A large portion of the MSc degree is attributed to the dissertation project. The aim of the project is to research and to evaluate the ability of automated systems to discover information about named entities and to incorporate this information in a knowledge source. The overall project involves two tasks: Entity Linking (EL), where names must be aligned to entities in the Knowledge Base, and Slot Filling (SF), which involves mining information about entities from text. SF can also be viewed as more traditional Information Extraction, or alternatively, as a Question Answering (QA) task, where the questions are static but the targets change.
I was involved in the Named Entity Recognition (NER) task, which is a part of the SF task. My responsibilities included developing or customizing an available NER system for the SF task, which involved conducting a research on the available tools for NER tagging, generate training data set for the NER, and generate models with the training data set and evaluating the tagged set based on the potential slot values tagged. I used Stanford NER system after analyzing the trade off with the other NER systems available. I generated training data for the Stanford NER system with the MUC 6 & MUC 7 datasets and Wikipedia text for 11 classes such as PERSON, ORGANIZATION, DATE, TIME, MONEY, NUMBER, COUNTRY, CITY, STATE, RELIGION and CAUSE OF DEATH. The training data set was used to train the classifiers. The trained classifiers were then used to classify and tag/annotate the given text with the 11 classes. Further, I developed a co-reference resolution system which involved tagging the pronominal references to the named entities. Both NER and co-reference resolution systems were used together for identifying the relations for which supervised learning approach was followed. The supervised learning approached involved training a Naive-Bayes classifier to generate whether a pair of named entities is involved in a relation. On identifying the relation between the entities, a list of probable slot values was produced.
The right slot values for each slot were selected and produced in the output format mentioned by the TAC. I used JAVA and Python to implement the mentioned tasks. Currently, the evaluation of each of these tasks is in progress.
Darwin Project Research: ‘Conversational speech to Clear speech’ for WIDEX
Darwin is a group project chosen during the course at 'The University of Sheffield', which provides an inevitable experience to work on research project. I opted to research on improving the speech intelligibility in the hearing aids for hearing impaired, titled 'Conclear: Enhancement of Speech Intelligibility', which was proposed by a Danish researcher from 'Widex'. We implemented the project in MATLAB, which involved developing a GMM (Gaussian Mixture Model) classifier to identify the vowels and the lexical stress from the speech. On identifying the vowels and the lexical stress from the speech, the vowel regions and the stress regions were modified to get two different results. I was involved in the identification and modification of the lexical stress. The identification of the lexical stress regions from the human speech involved training the GMM classifier using the annotated conversational speech data from TIMIT corpus. The modification of the speech involved exaggerating the identified stress regions based on a distance metric. Further, I was involved in the testing of the results, which involved subjective testing and objective testing. Also, a poster was presented on the project.
Information Retrieval System: Ranked Retrieval System
I was tasked with designing and implementing a document retrieval system, based on the ‘Vector Space Model’. For the implementation of the document retrieval system, Inverted Index was created after pre-processing the provided documents. Initially a simple Boolean retrieval method was used to retrieve the documents based on queries, then was extensively implemented using the vector space model by calculating the required values. The output was then tested on comparison with the CACM gold standard.
Research Associate at Evalueserve SEZ (Gurgaon) Pvt Ltd., (January 21, 2008 to June 30, 2010)
· Worked on various projects involving patent drafting, drafting defensive publications, patent-ability / novelty searches, patent invalidation / validity Searches, enforcement/ EoU/ claim chart analysis, patent portfolio analysis and patent landscape analysis in the domains of computer science, electronics, wireless technologies and encoding technologies. The responsibilities include understanding the client requirements, managing a team of professionals and delivering quality results within time-lines
· Handled various projects on competitive intelligent patent landscape studies, which requires in-depth analysis of patent portfolios of the client's competitors. The responsibilities include understanding client requirements, conducting initial research on client’s and competitors product and patent portfolio for benchmarking the study, analysing the patent portfolio of the client and the competitor, and delivering quality results with graphs and insights within the stipulated time frame
· Worked with various research firms on different technologies such as speech processing, video rendering, information retrieval and video fingerprinting
SKILLS:
Professional skills:
I have the ability to perform a requirement intake and develop an approach to resolve the problem/requirement statement. I am a conscientious go-getter who is highly organized, dedicated, and committed to professionalism. Further, I have analytical thinking with demonstrated talent for identifying, scrutinizing, improving, and streamlining complex work processes, which enabled me to come up with innovative solutions to solve complex technical problems. Moreover, I am a flexible team player who thrives in environments requiring ability to effectively prioritize and juggle multiple concurrent projects. I have been a results-driven achiever with exemplary planning and organizational skills, along with a high degree of detail orientation.
Management skills:
· Team management:
I am a goal-driven leader, maintain a productive climate and confidently motivate, mobilize, and mentor employees to meet high performance standards. I have managed teams of 2-4 professionals while working with Evalueserve. As a team lead, I brought out and utilized the functional talents in the team and delivered quality results to the clients. Further, I have been required to liaise with other teams, meet with the client, attend progress meetings, deal with issues in the team and make key decisions pertaining to the direction of the project. On the whole, I am a resourceful team player who excels at building trusting relationships with customers and colleagues.
· Time management:
As a part of the team and being a person leading the team, I have handled crucial deadlines for unstructured projects. I had come up with innovative solutions to manage time effectively as a team, which helped me meet the client expectations and receive accolades from the management.
Communication skills:
I am an exceptional listener and communicator who effectively convey information verbally and in writing. During the course of every project, I have communicated with clients from various English speaking countries which involved understanding the client requirements, business development and interaction with the client regarding the deliverables and feedback. Further, being the student and student-faculty representative, I have always voiced colleagues’ feedback and opinions in the student management and student-faculty meetings, toning my communication abilities further.
EXTRA-CURRICULAR ACTIVITIES
Student representative for MSc in computer science with speech and language processing course and Student-faculty representative for MSc computer science at ‘The University of Sheffield’. The responsibilities involve interacting with students, finding out the problem areas, analyzing and solving the problems by representing them before the staff committee.
PERSONAL DETAILS
- Languages known: English, Tamil, Kannada and Hindi
- Nationality: Indian (Entitled to work in the UK for 2 years after graduation under the Post Study Work (PSW) scheme)
Referees available on request
