Data Mining Research Proposal Summary  PhD in Information Systems Privacy preserving data

							Summary

     PhD in Information Systems (Privacy preserving data mining), Post Grad.
      Diploma in Management (Information Systems, Economics, Marketing), B.E.
      Electronics and Telecommunication.
     4+ years of research experience
     2 years of teaching experience
     1+ year of industry experience
     4 accepted journal papers
     3 accepted conference papers 1 under review
     Participant, NSF medium research grant proposal on Privacy Preserving Data
      Mining for distance based mining.
     Summer research experience, automation project for Maryland Voter
      Information Clearinghouse (http://mdelections.umbc.edu).
     Industry internship experience: PwCGlobal Consultants, on design and
      feasibility assessment of remote utility metering using hand held devices and SQL
      server interfacing.
     Industry experience: Business Development Manager (enterprise integration),
      Deloitte & Touche Consulting India Pvt. Ltd., Application consultant SAP BI7.0
      at IBM India Pvt. Ltd.



Academic Qualifications

     PhD (Information Systems), University of Maryland Baltimore County, USA.
      May 2007. (CGPA: 4.0/4.0)

         o Thesis Objective: - Developing a framework for data reduction and
           privacy preserving data mining for distance based algorithms.
               The objective was to build up a general framework for data
                  reduction and privacy preserving mining with distance based
                  mining algorithms. The framework encompasses a number of
                  algorithms/ heuristics with a general basis of orthogonally
                  transforming the data and doing some additional random
                  encryption post operations to further secure data. A provable
                  privacy guarantee of the schemes through a worst case measure
                  called amplification was further provided in some cases.
                  Computationally much less expensive extensions with high data
                  compression ratios were suggested as alternatives for special types
                  of data like time series and images, to be securely mined on
                  resource constrained and loss/theft-prone mobile devices. A
                  workable prototype system for the same was implemented and
                     evaluated for demonstration purposes on a commercially available
                     PDA phone.

      PGDM (Post Grad. Diploma in Mgmt.), Indian Institute of Management
       Calcutta, India 2003.

           o Major:- Management Information Systems
           o Minor:- Economics, Marketing

      B.E. (1st Class Hons.), Electronics and Telecommunication, Jadavpur University,
       India 2001.



Professional Experience


Internship

      Summer Intern as a Management Trainee at PriceWaterhouseCoopers Global
       June 2002-August 2002.

           o Design, analysis and assessment of the technical and economic feasibility
             of a scheme for Utility Billing and remote metering using hand held
             devices with SQL server interfacing.



Training and Certifications

      Completed 5 weeks training for SAP BI 7.0 consultancy ( TBW-10, 20, 41, 42,
       45) from SAP Academy

      SAP Academy certified as BI 7.0 Solution Consultant

      Participated in in-house training material preparation for SAP BI 7.0 at IBM India
       Pvt. Ltd. Center of Excellence.




Industry

      Application Consultant, SAP BI7.0, Global Delivery Services, IBM India Pvt.
       Ltd (June 2007-May 2007)
       o Unilever iFinance (distribution sector) – Responsibilities included
         development and QA check of developed objects from offshore,
         coordination of the offshore development team

                 Extensive experience in entire range of BEx reporting tools.
                  Challenges included using the enhanced features of the reporting
                  tools to meet complex reporting logic and strict formatting
                  specifications from client
                 Design and development of customized MS Excel based tools for
                  data generation. Data loading to BI from flat file sources of legacy
                  systems. Challenges included automating the generation of huge
                  volumes of clean and controlled de-normalized data for unit testing
                  of developed reports.
                 Backend design and development of cubes, data store objects and
                  data transfer processes. Challenges posed were complexity in
                  business logic and data interaction between different systems



       o AMGEN Global Enterprise Dashboard (distribution sector) –
         Responsibilities included development and QA check of developed objects
         from offshore, coordination of the offshore development team. Interaction
         with the client and onsite dash boarding team to obtain requirement
         specifications


                 Blueprinting and development of data flow from BEx queries to
                  direct update data store objects through use of Analysis Process
                  Designer. Challenges included unexplored nature of the APD tool
                  and how to incorporate data mapping from complex queries to data
                  store objects.
                 Use of Analysis Process Designer for generating data mining
                  models


   Business Development Manager (Enterprise Integration), Deloitte & Touche
    Consulting India Pvt. Ltd. (May 2007 - Present)

       o ITC Ltd (FMCG manufacturing and distribution sector) – Responsibilities
         included proposal preparation and engagement management for a
         Hyperion based financial planning solution implementation project
       o Union Bank of India (Finance & banking sector) – Development of RFP
         and successive proposal for a business intelligence strategy roadmap.
         Participation in cost structure composition and reverse auction bidding
         strategies for project acquisition
           o Mastek India Pvt. Ltd. (IT services sector) – Proposal development,
             engagement management and subsequent Project management after
             project acquisition for an end to end SAP BI 7.0 installation for FICO and
             HR modules.
           o Spencer’s Retail (Retail distribution sector) – Proposal development and
             prototyping of a weather and other supplementary data based data mining
             dashboard for merchandise demand and sales forecasting.


Research

     Research Assistant, Database group, Information Systems Dept. UMBC, (August
      2005-June 2006) and (August 2006-Present)

           o Near-optimal algorithms for maximizing accuracy of sanitized shared data
             while blocking mining of sensitive frequent item-sets.

           o Fourier transform based data compression and encryption frameworks for
             privacy preserving data mining with distance-based algorithms.

           o Fuzzy optimization modeling for coefficient selection for compression and
             encryption of very large datasets

           o Provable privacy preserving framework for k-nearest neighbor
             classification with orthogonally projected and additively perturbed data.

           o Enhanced local storage based secure framework for forensic image
             recognition and classification on hand-held devices.



Teaching

     Teaching Assistant ( August 2003-June 2003) and (August 2004-June 2004)

           o Responsible for preparing teaching materials on Knowledge
             Management for faculty lectures in Management Science, University of
             Texas at Dallas, USA and International School of Business Hyderabad,
             India.
           o Instructor for advanced level programming language course 247V,
             Visual Basic.NET for undergraduate Engineering Majors, Dept of
             Information Systems, University of Maryland Baltimore County.
Application Design and Development

      Research Assistant in Maryland Voter Information Clearinghouse project, a
       joint venture project between UMBC Dept. of Information Systems, Public Policy
       Research and Maryland State Board of Elections. June 2006-August 2006
       http://mdelections.umbc.edu

          o Requirement analysis, design and case based development platform
            selection/ evaluation for system design
          o Rapid application development of the online candidate registration and
            candidate search modules. The system was coded in PHP with Oracle 10g
            at the backend. Major challenges were to cope up with the widely varying
            user requirements, the sensitivity and irregularities of the data and the
            short development time.




Publications


Journal Papers

      Maximizing Accuracy of Shared Databases when Concealing Sensitive Patterns.
       Syam Menon, Sumit Sarkar, Shibnath Mukherjee. Information Systems
       Research Vol. 16, No. 3, pp. 256–270, September 2005
      A privacy-preserving technique for Euclidean distance-based mining algorithms
       using Fourier-related transforms. Shibnath Mukherjee, Zhiyuan Chen, Aryya
       Gangopadhyay. VLDB International Journal, Vol. 15, No. 3, pp. 293-315,
       September 2006
      A Fuzzy Programming Approach for Data Reduction and Privacy in Distance
       Based Mining. Shibnath Mukherjee, Zhiyuan Chen, Aryya Gangopadhyay.
       International Journal of Information and Computer Security Vol. 2, No. 1, pp. 27-
       47, 2008
      A privacy preserving technique for distance-based classification with worst case
       privacy guarantees. Shibnath Mukherjee, Madhushri Banerjee, Zhiyuan Chen,
       Aryya Gangopadhyay. Data and Knowledge Engineering Vol. 66, No. 2, pp. 264-
       288, 2008
      SAFROM: A Framework for Secure Automatic Face Recognition On Mobile-
       devices. Shibnath Mukherjee, Zhiyuan Chen, Aryya Gangopadhyay, Stephen
       Russell. Paper under review at a tier I conference on data security
Conference Papers

      A Modified Kangaroo Model for Long Lived Transactions Over Mobile
       Networks. Shibnath Mukherjee. Accepted Paper, IEEE region III
       SOUTHEASTCON 2003, Ocho Rios, Jamaica, April 4-6, 2003
      A Probabilistic Model for Optimal Searching of the Deep Web. Shibnath
       Mukherjee. Accepted Paper, Fourth International Conference on Intelligent Data
       Engineering and Automated Learning (IDEAL2003), Hong Kong, March 21-23,
       2003
      An Optimization Model for Planning and Utilization of Emergency Health Care
       Services Using Geographic Information Systems.         Shibnath Mukherjee.
       Accepted Paper for Poster Presentation, Fourth International Conference on
       Intelligent Data Engineering and Automated Learning (IDEAL2003), Hong Kong,
       March 21-23, 2003




Grant Proposals

      Participant/ alumni in National Science Foundation Medium grant proposal, IIS-
       IPS-0713345: A Privacy Preserving Framework for Distance-based Mining
       (09/01/2007-08/31/2010)


Software Skills

      Programming Languages: C, C++, VB.Net, Visual Basic 6
      Server side scripting Languages: PHP, ASP.Net
      Databases: Oracle 9i and 10g, MS SQL Server, MS Access
      ETL and reporting/ analysis tools: SAP BI7.0, IBM Intelligent Miner, Oracle
       Data Miner
      Mathematical Packages: Matlab 7.0 (special experience with signal processing,
       statistical and image processing toolboxes)
      Operating Systems: Dos, Windows 95, 98, 2000, XP
      General Purpose Packages: MS Office (experience with macro coding for all
       versions), Latex Text editor