Parag Tamhankar
Lecturer Computer Science- Education
B.Sc., Computer Science, North Maharashtra University, India, 1997
M.Sc., Computer Science, North Maharashtra University, India, 1999
Ph.D., Computer Science, Kavayitri Bahinabai Chaudhari North Maharashtra University, India, 2021
- Specializations
Programming language paradigms, operating systems, data structures and algorithm design, deep learning, pattern recognition, natural language processing, text mining.
- Biography
I joined Georgia State University as a faculty member on August 1, 2023. My doctoral research topic was “Creation of a Framework for Recognition and Transliteration of Offline Handwritten Historical Marathi Documents written in MODI Script into Devanagari Script.” During my research work, I designed and implemented several algorithms that were later published in international and Scopus-indexed journals.
Currently, I am working on improving the accuracy of the overall system by exploiting other alternatives, including using deep learning for tokenization of a document into lines and lines into characters. Moreover, I am exploring the possibility of giving annotations to the extracted and/or source data, which is required for training NLP algorithms. In the future, I want to gain deeper knowledge of NLP in order to explore the possibility of using it for semantic analysis to enhance my existing research work.
- Publications
International Journals (Published): # 03
- Papers Presented at International Conference: # 03
- Paper Presented at National Conference: # 03
Publications
- Tamhankar, Parag A., Krishnat D. Masalkar., and Kolhe S.R. "A Novel Approach for Character Segmentation of Offline Handwritten Marathi Documents written in MODI Script." ELSEVIER Procedia Computer Science 171 (2020): 179-187. (Scopus)
- Tamhankar P. A., Masalkar K.D., and Kolhe S.R. "Segmentation Of Overlapping, Skewed, And Touching Lines In Handwritten Modi Script Documents." Int J Comput Sci Appl, 14(2). Vol. 14, No. 2, ISSN 0972-9038. (Scopus)
- Parag Anil Tamhankar, Satish Ramesh Kolhe, "Skew detection, skew correction, and shirorekha extraction of line segmented offline handwritten historical marathi documents written in modi script", https://www.ijltet.org/journal_details.php?id=928&j_id=4531, Volume 10 Issue 1 - March 2018, 344-348, ISSN 2278-621X.
- Tamhankar P. A., Masalkar K.D., and Kolhe S.R. "Character Recognition of Offline Handwritten Marathi Documents written in MODI Script using Deep Learning Convolutional Neural Network Model" has been accepted for oral presentation at 3rd International Conference on Recent Trends in Image Processing and Pattern Recognition (RTIP2R, 2020) organized by Department of Computer Science & Information Technology in association with RUSA Center for Advance Sensor Technology and in collaboration with University of South Dakota, USA.
- Tamhankar P.A. and Kolhe S.R., "A New Valley Selection Criterion Based on Minimizing the Segmentation Error for Overlapping and Touching Lines in Handwritten MODI Script Documents", presented at a National Level Conference on Advances in Computing, School of Computer Sciences, NMU, March, 2017.
- Presented poster titled 'CHALLENGES IN RECOGNITION OF TYPED AND HANDWRITTEN MARATHI DOCUMENTS USING MODI SCRIPT' at University Level Avishkar March 14-15 2015, NMU. (2nd Prize)