Saurav Majumder

Email: saurav.majumder@ufl.edu     Phone: +1 (813) 600-0697

Objective
Seeking full-time Software Engineer position where I can use my programming skills along with my independent research abilities in solving challenging problems to help me and the company achieved their desired goal.

Education
Master of Science in Computer Engineering - December 2013
University of Florida, Gainesville, FL
Bachelor of Technology in Information Technology - May 2011
West Bengal University of Technology, Kolkata, India

Experience
Research Assistant, University of South Florida, Tampa Florida
January 2012 - December 2012
  • Researched on image segmentation algorithms, image features selection and classification algorithms for the detection of lung and sarcoma cancer from low-dose CT scan images.
  • Worked in collaboration with H. Lee Moffitt Cancer Center and sponsored by National Institute of Health.
  • Submitted a research paper in SPIE Medical Imaging 2013.
Teaching Assistant, University of South Florida, Tampa Florida
August 2011 - December 2011
  • Graded projects & homework assignments, proctored exams in Object Oriented Design course containing 56 students.
  • Assisted students in learning C++ programming through their class projects.

Skills
Certifications Sun Certified Java Programmer
Programming Languages C, C++, Java, SQL, HTML, CSS, PHP, Matlab
Tools Oracle, MySQL, Definiens Developer XD, MITK, VolView, MIPAV, FIJI, Weka, libSVM
Familiar With Hadoop, MapReduce, Amazon Web Services, Android Gesture API, Scala, JavaScript, JQuery, AJAX, Visual Basic, Twitter Bootstrap API, Google Places API, WAMP

Projects
  1. Effect of Texture Feature in Computer Aided Diagnosis of Pulmonary Nodules in Low-Dose CT scan.

    Studying the effect of texture feature in prediction of pulmonary nodules in Low-Dose Computed Tomography (CT). For this experiment 172 texture feature and 44 other features were used for classification using C4.5 Decision Tree, Random Forest, 5-Nearest Neighbor, Multinomial Logistic Regression and Support Vector Machine (SVM). Feature selection was performed using Recursive Elimination of Features (Relief-F), Wrappers with Best-First forward selection and Principle Component Analysi (PCA). After evaluating classifiers through 10-fold cross-validation (CV) and Leave-On-Volume-Out cross-validation it was observed that texture features have positive effect on nodule diagnosis.

  2. Developed classifier model for sarcoma cancer in CT images.

    Developing classifier model for the prediction of malignancy of sarcoma cancer from CT-scan images. My contribution to this project was in the process of proper segmentation of cancerous region using different clustering techniques for better feature extraction and prediction.

  3. Peer-to-Peer (P2P) File Transfer Application

    Peer-to-Peer file sharing software like BitTorent with some interesting features like choking-unchoking mechanism. The application was tested on 6 Linux machines sharing files of size 200 KB to 2 MB.

  4. Link Prediction in Marvel Universe Social Graph.

    Prediction of hidden links between comic characters in Marvel Universe Social Graph. The predictive model was designed with 9 aggregated and 3 topological features extracted from 12,507 comic characters and 37,083 comic books which amount to 99.1% of all the Marvel comic database. 7 popular classification algorithms: Bagging, J48, Random Forest, Naďve Bayes, JRip, k- Nearest Neighbors and Multilayer Perceptron were used. Random Forest showed the best result of 97.91% when classified 20,000 instances.

  5. Pattern recognition framework which includes implementation of multiple feature selectors, Bayes classifiers and k-NN.
  6. Websites: Wizlyst.com, Flickbazar.com and Jhatkadeal.com
  7. Hybrid Android to-do list with Click & Gesture (Download .APK)
  8. User-friendly chat client with testing tool
  9. Gator Restro - Database Management Systems
  10. yourCookBook - Human Computer Interaction

Other Activities
  1. Received Academic Achievement Award at University of Florida, 2013.
  2. National-level finalist in Web Designing category at World Skill Competition 2011, India.
    Three participants from all-over India reached the national finals in Web Designing, where we had to develop a complete web-application in 24 hours. The event was organized by NASSCOM and National Skill Development Corporation (NSDC).
  3. Finalist in ‘Service Innovation Challenge’ at 2nd International Conference on Services in Emerging Markets, 2011.
    My concept ‘Advanced Mobile Content Storage Web Application’ was selected for final round of development with IBM.
  4. Participated in IEEE Xtreme 5.0 and Facebook Programming Challenge.

Last Updated: September 25, 2013