Career Objective: Currently aspiring to be a doctoral student.
Education:
MS
New
Jersey Institute of Technology,
Computer Engineering, Dec 05.
BE
Graduated First Class with Distinction.
Computer Engineering, May 03.
GPA: 3.9 GPA in Undergraduate. 3.9 GPA in Master%u2019s.
Key Research fields: Neural
networks, Fuzzy Logic, Linux, Operating Systems development, Information
Retrieval using search engines, Graphics using OpenGL, Extreme programming in
Perl/PHP/C.
Research Interests: In Neural Networks, Pattern Recognition & Fuzzy logic (During M.S.)
1. Under Dr.Atam Dhawan, I worked my thesis on proposing and developing a neural network based algorithm called Topology-Based Fuzzy Clustering(TFC). This Algorithm has advantages of lesser execution time, accuracy and learning abnormal shapes compared to comparative algorithms.
2. Also worked on ALEC (Adaptive Learning and Clustering) developed at the Signal Processing lab headed by Dr.Dhawan.
3. Under Dr.Mengchu Zhou, worked on the Fuzzy Reasoning Petri Net (FRPN) model to assess the feasibility of using FRPN model in assessment of sensor fault-tolerance for an autonomous car.
4. Under Dr.Azzam-Ul Asar, a visiting faculty at NJIT, collaborated on a program, and future paper, that demonstrates the usefulness of applying Fuzzy logic (Takagi Sugeno Kang system) to MIL-882 probability standard.
5. Under Dr.Yehoshua Perl, I worked on developing a simulation model for a Worm Detection System using k-means clustering for the detection. This project was done for Telcordia Technologies.
Other Significant
Research experiences:
Graduate:
1. Under Dr.Yehoshua Perl, I worked on developing a better front-end for a semantic web project and modified the original back-end to support servlets.
2. Developed a web-based front end for a fuzzy logic based pattern matching program for IMS-Health.
3. A fully committed member of the NJIT Grandchallenge team (Oct 04-till day) participating in the annual DARPA Grandchallenge competition.
i.Major tasks done for the team include; handling IBM JS20 servers and developing a master control program that controls the operation of all the machines.
ii.Other tasks include but not limited to; setting & troubleshooting the NML (Neutral Messaging Library) backend system, migrating the servers and the other machine to Real-time Linux and programming concept programs for the sub-teams.
iii.Adapted Ant Colony Optimization for Path Planning. Also developed Kalman filter when differential GPS was unavailable.
3. Implemented a Kalman-filter based tracking program for moving objects as a requirement for CIS-780 under Dr.Chengjun Liu.
Under-Graduate:
1. Developed a simple protected-mode Operating System for the x86 Architecture. It featured a 32 bit address space, its own command interpreter and a Round Robin scheduler based kernel. Developed as a requirement for the %u2018Small Development Project%u2019 Course.
2. Industrial project for designing a prototype system to accurately measure the effect of Headlights in simulated real-life like conditions using OpenGL and VC (~14,000 lines). The project used 2D & 3D Texturing, Fogging, Mipmapping and other features of OpenGL.
3. Text based parser for correctly guessing the grammar of sentences. The rules were written in Yacc/Lex and the project was done as a course requirement for Language Processing.
4. An Interactive MFC based advanced version of Paint utility for Windows.
Computer Skills: Languages adept in: C, C , Matlab, Visual Basic 6.0, Visual C 6.0, Pascal, PERL, Tcl/Tk, Java, Prolog, HTML, XML, Assembly Language in windows and Linux ( x86/ 8051), Win32 programming, Shell scripting, PHP.
Development Environment: MS-Visual C 6.0, Dev Studio, Forte for Java%u2122.
API%u2019s known: Linux ASM, Win32, OpenGL, MFC, Working Knowledge in RMI, JINI and COM.
System Administration on: MS-DOS, Windows 95, 98, NT, 2000Pro, 2000Server, XP, Linux (Debian, Red Hat & Mandrake), and Windows Server 2003.
Network Skills:
1.Comfortable with managing and administrating networks on Windows and Linux.
2.Knowledge of Cluster computing on Linux.
Awards &
Honors:
1. Topped the Highschool & District in both the 10th and 12th Grades. Awarded the Dhirubhai Ambani Excellence Award on both occasions.
2.
Invited to visit Indian Space Research Organization (ISRO)
sites. One of the 14 students in
3. Awarded the Provost fellowship for graduate studies at NJIT.
4. Member of Tau Beta Pi. Engineering Honor Society
5. Invited to be member of Alpha Epsilon Lambda. Graduate Honor Society.
Training & Employment:
1. Internee at L&T e-Engineering center at
2. Worked as an Assistant to the Administrator of CS department at NJIT for a period of 1 year from January 2004 to Dec-2005 under Wenping Yang (wenping.yang@njit.edu).
Publications: In Progress:
1. Online Clustering with Single-pass Topology-Based Fuzzy Clustering
(TFC) Algorithm.
-Regular journal paper, in draft stage, available at
http://web.njit.edu/~ja44/paper.pdf
2. Using Fuzzy membership functions for MIL-882 standard
-Coauthoring with Dr.Asar, in draft stage.
Key Courses done: Data Structures and Algorithms, Operating Systems, Software Engineering, Database Management System Design, Computer Organization, Artificial Intelligence, Robotics, Object Oriented Programming, Computer Networks and Communication, Advanced Computer Architecture, Formal Languages, Computer vision, Neural Networks & Fuzzy logic.
Technical Work : Nominated and served successfully as the head of the %u201CTechnical committee%u201D that organized %u201CCommuniqué 2002%u201D, a national level event at DDIT, Nadiad. Managed all the events computer networks and solved problems where they had arisen.
Current Work: Writing tutorials and explanations on Computer programming languages (Perl, PHP, VC ), Kalman filters, Fuzzy logic, neural networks, hardware and online programs like RSS feeders, Regular Expression extractors; and other web based utilities.
My current work can be seen at ( http://web.njit.edu/~ja44 )
Master%u2019s Thesis Abstract:
Online clustering is of significant interest for real-time data analysis. Generic offline clustering methods such as K-Means, C-Means and others are computationally expensive. The computational burden of these methods increases non-linearly with the size of the data set. In addition these methods usually require a good amount of supervised knowledge yielding a non-unique solution. For real-time data analysis, there is an important tradeoff between accuracy and computational efficiency. An unsupervised one-pass clustering method that efficiently adapts to data distribution and evaluation is proposed. This method, Topology-Based Fuzzy Clustering (TFC), uses the topology of data to discover clusters. TFC uses the method of Growing Neural Gas (GNG) method of creating linked sub-clusters and extends GNG by assigning a fuzzy membership to the sub-clusters, noting the link structure for creating clusters and influencing the learning nodes at each sub-clusters. This also gives a fuzzy estimation of data distribution within each cluster. The computational burden for TFC is proportional to the size of the initial data set and increases linearly with the addition of new data.
As TFC is based partly on GNG, it is an unsupervised algorithm. A supervised learning method is proposed that can be used in conjunction with TFC, to increases its accuracy with minimum computational burden. This adaptive algorithm is called the Adaptive Topology-Based Fuzzy Clustering (ATFC). In this study, the performance of ATFC and TFC is also evaluated against standard datasets.
The full thesis is available at http://web.njit.edu/~ja44/TFCThesis.pdf