Knowledge is skill in action.

Learning is an evolution.

Premjith B

I am a PhD student at Center for Computational Engineering and Networking at Amrita Vishwa Vidyapeetham, Coimbatore, India under the supervision of Dr. K.P Soman. My current research focus is on Natural Language Processing (NLP) and Deep learning. I am interested in applying deep learning techniques in the computational processing of Indian languages. My dissertation is on Embedding linguistic features for improving Neural Machine Translation from English to Indian languages, particularly Malayalam and building deep learning based NLP tools for Malayalam. Apart from Malayalam, I also developed deep learning based NLP tools for various Indian languages. I also worked on applying kernel methods and explicit feature mapping algorithms in predicting network anomalies. Now, I am learning the maths of Reinforcement Learning and Dynamic Mode Decomposition (DMD).


My research focuses on Natural Language Processing - in particular, I am working on the development of NLP tools for linguistically rich Indian languages using deep learning algorithms. These tools can be effectively utilized in Neural Machine Translation, Indian language spoken dialogue system, Robotics, social media text analytics etc. Along with the machine learning/deep learning tools, I also prepared data-sets for different NLP tasks such as Machine Translation, Morphological analysis, Sandhi splitting, Parts-of-Speech tagging and Named Entity Recognition in Indian languages.

Goal of my research is to take the common man to the world of flooded knowledge and let them learn through their spoken language or mother tongue.

Some topics of interests are:

  • Neural Machine Translation
  • Incorporating Artificial Intelligence to learn the grammatical rules of natural languages.
  • Models of linguistically rich languages such as Sanskrit, Malayalam, Tamil, Hindi, Telugu and also for Arabic
    • Morphological analyser
    • Sandhi splitter
    • Named Entity Recognition (NER) tagger
    • Parts of Speech (POS) tagger
    • Word Sense Disambiguation (WSD)
  • Social media text analytics
  • Biomedical text mining
  • Kernel methods and explicit random feature mapping algorithms
  • Network anomaly detection
  • Deep learning - in particular, Recurrent Neural Network and Long Short-Term Memory Networks
  • Reinforcement Learning
  • Probability and Graphical Models

Collaborative research projects

  • Sanskrit NLP for Ayurveda text processing with Amrita School of Ayurveda.
    • Developed a Sanskrit POS tagged system. Submitted a paper at FIRE 2018
    • Developing morphological generators for Sanskrit nouns and verbs
    • Developing a dependency parser for Sanskrit
    • Word2vec representation for all the texts available in Digital Corpus of Sanskrit
  • Building a Malayalam Wordnet in association with Italian project on Universal Knowledge Core (UKC) with Trento University, Italy

Teaching Assistance

  • 16CN613 Deep Learning and Probabilistic Graphical Models
  • 16 MA603 Computational Linear Algebra for Data Sciences
  • 16CN604 Computational methods for Optimization
  • 17AL 601 Linear Algebra and Optimization for Signal Processing
  • 17AL 605 Deep Learning


Professional Experience

  • Research Assistant (October 2014 - )
    • Center for Computational Engineering and Networking, Amrita Vishwa Vidyapeetham, Coimbatore
  • Assistant Professor (August 2012 - September 2014)
    • Royal College of Engineering and Technology, Thrissur, Kerala
  • Lecturer (January - April 2010)
    • Viswa Jyothi College of Engineering and Technology, Muvattupuzha, Kerala


Google scholar profile Research gate profile LinkedIn Profile

  • Premjith B, M Anand Kumar, Soman K.P (2018). Neural Machine Translation system for English to Indian Languages translation using MTIL Parallel Corpus . In Journal of Intelligent Systems
  • Premjith B, Soman K.P, M Anand Kumar (2018). Deep learning based morphological analysis of Tamil nouns and verbs. In Research Conference on Data and Decision Science (RCDDS' 18)
  • Premjith B, Soman K.P, M Anand Kumar (2018). A deep learning approach for Malayalam morphological analysis at character level. In International Conference on Computational Intelligence and Data sciences (ICIDS 2018)
  • Aravind Jaya Prakash and Bhavukam Premjith Dhanya Sathyan, Kalpathy Balakrishnan Anand (2018). Modeling the Fresh and Hardened Stage Properties of Self-Compacting Concrete using Random Kitchen Sink Algorithm. In International Journal of Concrete Structures and Materials 12.1: 24.
  • Ratnam, D. J., Kumar, M. A., Premjith, B., Soman, K. P., & Rajendran, S. (2018). Sense Disambiguation of English Simple Prepositions in the Context of English–Hindi Machine Translation System. In Knowledge Computing and Its Applications (pp. 245-268). Springer, Singapore.
  • K P Soman R. Vinayakumar, S. Sachin Kumar, B. Premjith, & Poornachandran Prabaharan (2017). Deep Stance and Gender Detection in Tweets on Catalan Independence@Ibereval 2017. In Proceedings of the Second Workshop on Evaluation of Human Language Technologies for Iberian Languages (IberEval 2017)
  • R Vinayakumar, Premjith B, Sachin Kumar S, Prabaharan Poornachandran . (2017). deepCybErNet at EmoInt-2017: Deep Emotion Intensities in Tweets. In Proceedings of the 8th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis (pp. 259-263).
  • Aravind J Prakash, Dhanya Sathyan, K B Anand, Premjith B (2017), Prediction of rheological properties of self compacting concrete: Regularized least square approach. International Journal of Earth Sciences and Engineering
  • Vinayakumar, R., Kumar, S., Premjith, B., Prabaharan, P., & Soman, K. P. DEFT 2017-Texts Search@ TALN/RECITAL 2017: Deep Analysis of Opinion and Figurative language on Tweets in French. In 24e ConfĂ©rence sur le Traitement Automatique des Langues Naturelles (TALN) (p. 99).
  • Premjith, B., Kumar, S. S., Shyam, R., Kumar, M. A., & Soman, K. P. (2016). A Fast and Efficient Framework for Creating Parallel Corpus. Indian Journal of Science and Technology, 9(45).
  • Soman K.P Prabaharan Poornachandran, Premjith B (2016) . A distributed approach for predicting malicious activities in a network from a streaming data with support vector machine and explicit random feature mapping. The IIOAB Journal
  • Kumar, S. S., Premjith, B., Kumar, M. A., & Soman, K. P. (2015, December). AMRITA_CEN-NLP@ SAIL2015: Sentiment analysis in Indian Language using regularized least square approach with randomized feature learning. In International Conference on Mining Intelligence and Knowledge Exploration (pp. 671-683). Springer, Cham.
  • Premjith, B., & Soman, K. P. Computational Experiment of One Class SVM in Excel. International Journal of Applied Engineering Research 10 (20), 19356-19360
  • Premjith, B., Mohan, N., Poornachandran, P., & Soman, K. P. (2015). Audio Data Authentication with PMU Data and EWT. Procedia Technology, 21, 596-603.
  • Premjith, B., Kumar, S. S., Manikkoth, A., Bijeesh, T. V., & Soman, K. P. (2013). Insight into Primal Augmented Lagrangian Multilplier Method. arXiv preprint arXiv:1312.7637.
  • Premjith B. Vidya M. Poornima S .V. and K.P Soman. A Level Set Methodology for Sanskrit Document Binarization and Character Segmentation. (Best paper award)


Faculty Development Programme on Machine Learning at Ernad Knowledge City - Technical campus, Manjeri, Malappuram (18 July, 2018)

Recurrent Neural Network (RNN) and Long Short Term Memory (LSTM) networks and its applications at Summer course on AI and Data science at Amrita Vishwa Vidyapeetham, Coimbatore (24 May, 2018)

Weka hands-on at Summer course on AI and Data science at Amrita Vishwa Vidyapeetham, Coimbatore (21-22 May, 2018)

Support Vector Machine (SVM) and Deep Learning (Theory and hand-on) at Faculty Development Programme on Machine Learning at College of Engineering Thalassery (26 April 2018)

Recurrent Neural Network (RNN) and Long Short Term Memory (LSTM) networks and its application in Malayalam language processing at Faculty Development Programme on Machine Learning at, Vidya Academy of Science and Technology (20 January, 2018)

Recurrent Neural Network (RNN) and Long Short Term Memory (LSTM) networks in DeepChem 2017 workshop organized by Center for Computational Engineering and Networking (CEN) at Amrita Vishwa Vidyapeetham (22-24, December, 2017)

Neural Machine Translation and building a Neural Machine Architecture using Tensorflow in Machine Translation in Indian Languages - Shared task cum workshop 2017 at Amrita Vishwa Vidyapeetham (7 - 8, September, 2017)

A hands on session on using LaTex for writing research papers at College of Engineering, Thalassery

Workshop on arduino and raspberry pi at Rajagiri School of Engineering and Technology

Workshop on arduino and raspberry pi at Royal College of Engineering and Technology

Workshops co-organized at Center for Computational Engineering and Networking (CEN), Amrita Vishwa Vidyapeetham, Coimbatore

  • Summer course on AI and Data science (21st May 2018 - 1st June 2018)
  • Workshop on Data -Driven Modelling 2018 (8 -9 January, 2018 )
  • DeepChem 2017: Deep Learning & NLP for Computational Chemistry, Biology & Nano-materials (22-24 December, 2017)
  • A Refresher experiential course on linear algebra and Optimization for Most Modern Signal processing and pattern classification (25 - 27 November, 2017 )
  • DeepSci 2017 Workshop: Deep Learning for Healthcare and Financial Data Analytics (11 November, 2017)
  • AISec 2017 Workshop: Modern Artificial Intelligence (AI) and Natural Language Processing (NLP) Techniques for Cyber Security (28 October, 2017 )
  • Shared task cum workshop on Machine Translation in Indian Languages - MTIL 2017 (7-8 September, 2017)
  • Workshop on Web Application and Cyber Security (30 November - 4 December, 2015)
  • Workshop on High Performance Computing and Bigdata Analytics (19 September 2015)
  • Workshop on Formal Methods for Software Design and Verifications (10 January 2015)
  • Workshop on Big Data and Probabilistic Graphical Models (27-29 December 2014)
  • Workshop on Distributed Computing Algorithms (for machine learning) and Apache-Spark Framework for Big Data Analytics (30 October - 1 November, 2014)
  • Workshop on Sparse Image & Signal Processing (SISP -2011)

Technical skills

  • Skills
    • Machine learning, Deep learning, Probabilistic Graphical Model, Kernel methods, Explicit random feature mapping algorithms, Compressive sensing, Natural Language Processing, Computational processing in Indian languages
  • Programming languages
    • Python, Matlab, C, Octave, Julia, Weka
  • Deep learning / Machine learning tools and frameworks
    • Tensorflow, Keras, NLTK, Scikit-learn, Pandas, Gurls, LibSVM, Deep learning toolbox - Matlab, Matlab NLP-Master, cvx, cvxpy, OpenNMT
  • Operating system
    • Linux and Windows
  • Documentation tools
    • LaTex, Microsoft office, LibreOffice