Saurabh Kumar

From Msplacements

Url: http://research.iiit.ac.in/~saurabh
E-mail: saurabh@research.iiit.ac.in
Contact No: +91 9959609816

Career objective
Seeking a challenging, research and development career in Data Mining, wherein I can apply my knowledge and acquire new skills.

Research & Industrial Experience

• July 2007-Till Date: Research Assistant, Center for Data Engineering, IIIT.
• June 2006 – July 2007: Software Engineer, Verizon Data Services India, Hyderabad.
  
• July 2005- May 2006: Software Engineer, DBSentry Corp. Pune.

Education

• M.S. by research (CSE) , IIIT-Hyderabad, 2007-2009 (expected) 7.4/10
• B.E (CSE), Institute of Technology, G. G. Vishwavidyalaya, Bilaspur, 2001-2005, 76.13%

Publications

• Saurabh Kumar, Kamalakar Karlapalem: Subspace Outliers Analytics. Under Review.
• Amit S, Saurabh Kumar, Jaideep, Vasudev Verma: Towards Analyzing Data Security Risks in Cloud Computing Environments. In proceeding of International Conference on Information Systems, Technology, and Management (ICISTM 2010).

Primary Research & MS Thesis
Title : Outlier Mining in Subspaces
Advisor: Prof. Kamalakar Karlapalem

Research Interests

• Data Mining
• Web Mining
• Cloud Computing
• Information Retrieval
• Data Warehouse
• Pattern Recognition

Related Course work

• Data Warehousing and Data Mining
• Pattern Recognition
• Artificial Neural Network
• Soft Engg and Cloud Computing

Computer Skills

• Programming Languages: C, C++, Java, C#
• Scripting Languages: PHP, Perl
• Database: Oracle, SQL Server, MySQL
• Web Technologies: HTML, XML, CSS, JavaScript
• Operating Systems: Windows, Linux
• Tools and Libraries: WEKA, RapidMiner, Hadoop

Achievements

• Got Sun Certified Java Programmer Certificate
• GATE qualified with 97 percentile in 2007.
• Participated in National Level Industrial Trade Fair at Bilaspur.
• Got Honour Certificate from PT. Kali Charan Educational Society, U.P.

Projects

• Bayesian Classifier in Hadoop environment
  In this I implemented the Naïve Bayesian Classifier in Hadoop environment using Map-Reduce paradigm, so that we can run the classifier on different machines and can classify very large dataset very fast.
• Online clustering of Linguistic Data
  In this project, I implemented a single-pass text clustering algorithm designed specifically for clustering news stories and examine its empirical behavior. I then analyze some of its key design features and compare them to possible alternative methods.
• Verizon Iobi
  Verizon iobi is a innovative communications management software applications that allow users to manage their voice and data communications from wherever they are, any time of day. It is available in the form of messenger or a Web site access portal - or by landline or cell phone using a voice recognition portal. Verizon iobi includes real time call management. It provides services like route incoming calls to Voice Mail or another number in real time, schedule call forwarding in advance, send email and text messages, view and update an address book and calendar.
• Verizon FioS
  Verizon FioS provides Internet, TV, telephone over a single line operating over a fiber-optic communications network. I worked on FioS TV Portal. The site allows customers to use any desktop or laptop computer to review their local TV schedules and information about VOD (Video on demand) and pay-per-view offerings. Also allow customers to control their DVRs and set up and modify recordings remotely and manage parental controls etc.
• File Synchronizer (FileSync)
  FileSync is a robust cross platform file synchronizer, developed using the JXTA technology. It syncs network files, making it easy to move new or updated files and file permissions while keeping mirrored directories up-to-date. It is also a smart way to protect data in case of a disaster.