Dr. Rajen Bhatt

Since June-2005, working at Samsung India Software Centre, Noida, INDIA as a Lead Engineer.
Project Associate under NPTEL at Electrical Engineering Department, IIT Delhi, INDIA from Sept’04 – May’05.
Research Fellow at Electrical Engineering Department, IIT Delhi, INDIA from Dec’01 – May’05.
Visiting faculty, Electrical Engineering Department, Delhi College of Engineering, INDIA from Dec. 2000- Nov. 2001.

EDUCATION

Ph.D. (2005) Title of the Thesis – Fuzzy-Rough Approach to Pattern Classification: Hybrid Algorithms and Optimization.

Department of Electrical Engineering, IIT Delhi, INDIA.

M.E.(2002) Control and Instrumentation

Department of Electrical Engineering, Delhi College of Engineering, University of Delhi, Delhi, INDIA.

1^st Division, 77.25% marks, 1^st in order of merit.

B.E.(1999) Instrumentation and Control

S. S. Engineering College, Bhavnagar University, Bhavnagar, INDIA.

Distinction, 68.5% marks, 3^rd in order of merit.

AREAS OF RESEARCH

Soft Computing Tools and their Hybridizations
Data Mining
Decision Support Systems
Supervised and Unsupervised Learning
Reinforcement Learning

PROJECTS UNDERTAKEN

Lead Engineer,

Samsung India Software Centre (June’05 – Till date)

Automatic Image Redirection

Presently working on the development of software product for Automatic Image Redirection based on image symmetry analysis and classification techniques.

Environment: MATLAB

Role: Algorithm analyst and Lead engineer

Natural Color Expert for Visual Displays

Environment: Windows, Mac

Role: Algorithm analyst and Lead Engineer

Image segmentation by histogram thresholding based on fuzzy uncertainty measures

Environment: MATLAB, MATLAB compiler

Role: Principal Inventor and Lead Engineer

Project Associate, NPTEL, Electrical Engineering Department, IIT Delhi (Sept’04 – May’05)

Project: Web-based course for Control Engineering

Principal Inventor: Prof. M. Gopal, Electrical Engineering, IIT Delhi

Responsibilities: Development of interactive tutorials and exercises for MATLAB programming, SIMULINK, Control System Toolbox, SISO Design Tool, and LTI View.

Nature of the project: This is the funded project by MHRD under the NPTEL (National Program on Technology Enhanced Learning). The project involved development of web-based multimedia course for Control Engineering. Soon this complete course and the developed MATLAB modules will be published on web by MHRD.

PROFESSIONAL AFFILIATIONS

Reviewer, Pattern Recognition Letters, Elsevier Publications
Reviewer, Information Sciences, Elsevier Publications
Reviewer, Pattern Analysis and Applications, Springer-Verlag

PUBLICATIONS

JOURNALS

Rajen B. Bhatt and M. Gopal, 2006, “Neuro-Fuzzy Decision Trees”, International Journal of Neural Systems, 16 (1), pp. 63-78.

Rajen B. Bhatt and M. Gopal, 2006, “On the Extension of Functional Dependency Degree from Crisp to Fuzzy Partitions”, Pattern Recognition Letters, 27 (5), pp. 487-491.

Rajen B. Bhatt and M.Gopal, 2005, “Improved Feature Selection Algorithm with Fuzzy-Rough Sets on Compact Computational Domain”, International Journal of General Systems, 34 (4), pp. 485-505.

Rajen B. Bhatt and M. Gopal, 2005, “On the Compact Computational Domain of Fuzzy-Rough Sets”, Pattern Recognition Letters, 26 (11), pp. 1632-1640.

Rajen B. Bhatt and M.Gopal, 2005, “On Fuzzy-Rough Sets Approach to Feature Selection”, Pattern Recognition Letters, 26 (7), pp. 965-975.

Rajen B. Bhatt and M.Gopal, 2004, “On the Structure and Initial Parameter Identification of Gaussian RBF Networks”, International Journal of Neural Systems, 14 (6), pp. 373-380.

CONFERENCES

Rajen B. Bhatt, “Image Segmentation by Histogram Adaptive Fuzzification”, IEEE-INDICON 2005, Dec 11-13, Chennai, INDIA, 2005.

Rajen B. Bhatt and M.Gopal, “Improving Learning Accuracy of Fuzzy Decision Trees by Direct Back-propagation”, IEEE International Conference on Fuzzy Systems (IEEE-FUZZ’05), pages 761-766, May 23-26, Nevada, US, 2005.

Rajen B. Bhatt and M.Gopal, “FRID: Fuzzy-Rough Interactive Dichotomizers”, Proc. IEEE International Conference on Fuzzy Systems (IEEE-FUZZ’04), pages 1337-1342, July 26-29, Budapest, Hungary, 2004.

Rajen B. Bhatt and M.Gopal, “Induction of Weighted and Interpretable Fuzzy Classification Rules for Medical Informatics”, Proc. International Conference on Systemics, Cybernatics, and Informatics (ICSCI 2004), pages 371-376, Feb. 12-16, Hyderabad, INDIA, 2004.

Rajen B. Bhatt and M.Gopal, “Fast Fuzzy-Rough Hybrid Algorithm for Automatic Rule Induction”, Proc. International Conference on Recent Trends and New Directions of Research in Cybernatics and Systems Theory”, Jan. 1-3, Guwahati, INDIA, 2004.

Rajen B. Bhatt and M.Gopal, “Fuzzy Decision Trees with Gaussian Kernel for Bi-directional Neuro-Fuzzy Transformation”, Proc. Of 28^th National Systems Conference (NSC-2004), Dec. 16-18, Vellore, INDIA, 2004.

Shweta Sahu and Rajen B. Bhatt, “Automatic classification of Electroencephalography Signals using Wavelet Packet Analysis and Fuzzy Decision Trees”, Proc. of 28^th National Systems Conference (NSC-2004), Dec. 16-18, Vellore, INDIA, 2004.

Shweta Sahu, Rajen B. Bhatt, Jayashree Santhosh, Sneh Anand, “Transition in Translation Algorithms in the Brain-Computer Interface Systems: A State of the Art”, Proc. of 12^th National Convention of Electronics and Telecommunications Engineers, IETE Conference on Medical Electronics, December 2004.

Rajen B. Bhatt, “Neural Networks: An Introduction”, AICTE-STTP on MATLAB for Design and Analysis of Electrical, Electronics, and Control Systems, Nirma Institute of Technology, Ahmedabad, INDIA, Dec. 1-12, 2003.

SUBMITTED

Rajen B. Bhatt and M.Gopal, “FRCT:Fuzzy-Rough Classification Trees”, Pattern Recognition Letters.

INTERNATIONAL WORKSHOPS

CIMPA- UNESCO – INDIA School on Soft Computing Approach to Pattern Recognition and Image Processing

An International Workshop organized by Machine Intelligence Unit, Indian Statistical Institute, Kolkata, INDIA, Dec. 2-14, 2002.

SOFTWARES DEVELOPED

RSDA : A Toolbox for Rough Set Data Analysis

It includes core algorithms for performing rough set data analysis such as: dependency analysis, significance analysis, reduct calculation, tree-structured rule generation, and rule application.

§ FRILM: A Toolbox for Fuzzy-Rough Inductive Learning Methodologies

It includes algorithms for fuzzyfication through FCM algorithm, fuzzy cluster validity, and fuzzy decision tree generation through six referred algorithms; fuzzy ID3, Yuan and Shaw’s method, Yeung et.al. method, FRCT (Rajen & Gopal, 2004), FRID ver. 1.1 (Rajen & Gopal, 2004), and FRID ver. 1.2 (Rajen & Gopal, 2004). It also includes algorithms for min-max-max, product-product-sum, and weighted rule inference mechanism. Some other features are: Fuzzy-Rough feature selection algorithm (Rajen & Gopal, 2004), and fuzzy rule base reduction algorithm.

§ CLONALG: A Toolbox for Cloning Selection Algorithms for Optimization and Pattern Recognition.

§ NEURO-FDT: A Toolbox for direct back-propagation learning of fuzzy decision trees by keeping its hierarchical structure intact. Neuro-fuzzy decision trees have been proposed by Rajen & Gopal (2004). It also includes methodology for the identification of structure and initial parameters of Generalized Gaussian Radial Basis Function Networks (GRBFN) based on fuzzy decision trees (Rajen & Gopal, 2004).