Home

I am currently a graduate student at Language Technologies Institute (LTI), CMU.

I worked as a Research Assistant at CNGL, Dublin City University, Ireland with Dr. Antonio Toral, Dr. John Judge, Prof. Qun Liu and Prof. Josef van Genabith. I also worked with Dr. Declan Groves previously. My primary interests are Natural Language Processing (especially Dependency Parsing and Machine Translation) and Machine Learning. My current work in on trying to predict a better translation given multiple machine translation outputs having no access to the reference data set. I am also working on Quality Estimation of a translation outputted by an MT system.

I completed BTech and MS (Dual Degree) in Computer Science at IIIT-Hyderabad, India. I did my Masters in Natural Language Processing under the guidance of Prof. Rajeev Sangal and Mr. Prashanth Mannem. My thesis is mainly focused on transferring dependency relations from English to Indian Languages there by automatically building a treebank and a parser for it.





Research Publications

  1. Aswarth Dara , Josef van Genabith, Qun Liu, John Judge and Antonio Toral "Active Learning for Post-Editing Based Incrementally Retrained MT", In Proceedings of EACL 2014, Gothenburg, Sweden

  2. Aswarth Dara , Sandipan Dandapat, Declan Groves and Josef van Genabith "TMTprime: A Recommender System for MT and TM Integration", In Proceedings of Demo Session on NAACL-HLT 2013, Atlanta, GA, USA

  3. Puneeth Kukkadapu, Deepak Kumar Malladi and Aswarth Dara "Ensembling Various Dependency Parsers: Adopting Turbo Parser for Indian Languages", In MTPIL Workshop at COLING 2012, Mumbai, India

    (Achieved best accuracy in 4 out of 6 tracks)


  4. Aswarth Dara, Prashanth Mannem, Hemanth Sagar Bayyarapu and Avinesh PVS "Transferring Syntactic Relations from English to Hindi Using Alignments on Local Word Groups", In Proceedings of IJCNLP 2011, Chiang Mai, Thailand

  5. Prashanth Mannem and Aswarth Dara "Partial Parsing from Bitext Projections", In Proceedings of ACL-HLT 2011, Portland, OR, USA

  6. Aswarth Dara and Prashanth Mannem "Bidirectional Parser for Indian Languages", In Proceedings of ICON 2010 Tools Contest, Kharagpur, India

  7. Prashanth Mannem, Aswarth Dara and Akshar Bharathi "LTAG Spinal Treebank and Parser for Hindi", In Proceedings of ICON 2009, Hyderabad, India

Courses

Some of the Advanced courses I have taken in IIIT-Hyderabad include the following:
  • Machine Learning
  • Information Retrieval and Extraction
  • Pattern Recognition
  • Data Warehousing and Data Mining
  • Artificial Intelligence
  • Natural Language Processing
  • NLP Applications

Some of the Interesting courses I have taken in IIIT-Hyderabad include the following:
  • Introduction to Cognitive Science
  • Intelligence: A Technology of Mind
  • Social and Technical Innovation
  • Gandhi and India
  • Readings from Hindi Literature
  • Corporate Strategy (Audited)

Conferences and Workshops attended

Some of the conferences, workshops and data meetups which I have attended-

Major Projects

  1. TMTprime: A Recommender System for MT and TM Integration
    Given a set of multiple MT systems and a document to translate, the goal is to recommend the MT outputs in such a way that the recommended outputs should have the least post-editing effort.
    Advisors - Dr Declan Groves and Prof Josef van Genabith

  2. Partial Parsing from Parallel Texts
    Work involves training a statistical parser from projected annotations given a parallel corpus and a source dependency parser. The goal of the project is to develop resources for resource poor languages.
    Advisors - Mr. Prashanth Mannem and Prof. Rajeev Sangal

  3. Transferring syntactic relations using local word groups on alignments
    Work involves effectively transferring the syntactic relations from source to target language given the source side dependency parser and word aligned parallel corpora. It also investigates the im- portance of Local Word Groups both in word alignment and during projection. The experiments are conducted for English to Indian languages.
    Advisors - Mr. Prashanth Mannem and Prof. Rajeev Sangal

  4. Design of Wikipedia Search Engine
    Involves creating a fast and efficient document indexer and retrieval engine for Wikipedia corpus containing more than 7 Million pages. Efficient ranking, indexing and retrieval algorithms are used to make the search time less than 0.2 seconds. Advanced techniques like compressing, stemming, buffering etc. are used in making the engine faster.
    Advisor - Dr. Vasudeva Verma

  5. LTAG spinal Treebank and Parser for Hindi
    Work involves extracting LTAG spinal Treebank from Hyderabad Dependency Treebank for Hindi and training a statistical parser on it. The idea is to convert treebanks available in one formalism into another automatically resulting in saving significant effort and time.
    Advisor - Dr. Prashanth Mannem

  6. Minor Projects

    • Online Handwriting Recognition as part of a course project (Pattern Recognition)
    • Implementation of various classifiers line KNN, Neural Networks and Linear Discriminant Analysis
    • Simulation of Pacman Game in OpenGL (Computer Graphics)
    • Pruning Phrase Table in SMT using Local Word Groups (NLP Applications)
    • Simulated Linux shell with basic commands (Operating Systems)

Academic Honours

  • Research Award for Excellence in 2009-10

  • Dean's Merit List (Monsoon 2009)

  • Receipt of Prathibha scholarship from state government of Andhra Pradesh for 4 years of undergraduate

  • Awarded two travel grants from Microsoft Research India to present at ACL and IJCNLP conferences

  • Secured All India Rank of 1961 in AIEEE conducted throughout India

  • Secured a rank of 89 in EAMCET conducted throughout Andhra Pradesh

  • Abstract has been accepted at Unisys Cloud 20/20 Student Technical Paper Contest 2010

Academic Experience

  • Research Assistant at CNGL, Dublin City University, Ireland    June'12-May'14
  • Associate Software Engineer at CA Technologies, Hyderabad, India     Jul'11 -Dec'11
  • Teaching Assistant for Cloud Computing     Aug'10 - Dec'10
  • Teaching Assistant for NLP Applications     Dec'09 - May'10
  • Research Assistant at Language Technologies Research Center, IIIT Hyderabad     Aug'09-July'11

About Myself

I strongly believe the quote "Follow your passion, the rest will follow you." Whenever I am sad about something, this quote gives me a lot of energy to get back to normal with out worrying too much.

I made a new hobby of playing Badminton these days at regular intervals and of course my game is improving a lot. I am a vegetarian food lover and always try to find new restaurants in Dublin that suits the same. I like watching English TV shows, hanging out with friends and going for movies whenever I get some free time. I am also interested in visiting places as it provides some recreation sidelining from the daily research work.

I love attending conferences as they provide a platform to interact with various researchers across all over the world and also to get the feedback for your own research work. Its also fun to visit other places while attending conferences. I attended ICON 2008, ICON 2009, ACL 2011, IJCNLP 2011, LxMLS 2012, LRC 2012, AMTA 2012, COLING 2012, NAACL 2013 and EAMT 2014 (probably) conferences so far and hope the list will increase much further.