New Course Announcement

(Under Construction)

 

ISM 209: Knowledge Services and Data Analytics

 

 

 

Professor Ram Akella

UCSC Silicon Valley Center (Bldg 19, Room 2089, NASA Research Park) & E2-563

     akella@soe.ucsc.edu,  Tel:  650-279-3078

 

 

 

 

When: Fall 2006, Tuesdays 6-9 pm  (1st Lecture: Tuesday, September 26)

Where: Silicon Valley Center, Classroom 2073, and live telecast to E2-475

 

Who :(should take this course?):

 

- Engineers, Managers, and Executives who would like acquire an understanding of the new management analytics and services required by exciting firms such as Google, Yahoo, Microsoft, IBM, HP, Cisco and Fair Isaacs. These analytics include data/text mining, optimized search and marketing/sales, enterprise strategy including new product development, financial engineering, knowledge services and support, and supply chain management

- Entrepreneurs: TIE (The Indus Entrepreneurs) entrepreneurial sessions and coaching available on request, in addition to regular course lectures

- All SOE and SSD-Economics students with strong analytic and business interests including those who wish to develop skills in some of the following areas and obtain attendant benefits:

 

-          Learn about data mining, machine learning, and analytic/mathematical approaches in business and technology management :

-          Search engines and marketing, new product development, supply chain management,  financial engineering, and technology and IT management

-          explore doing startups in these areas

-          start training to work in companies such as  Cisco, Yahoo and  Google in search and data mining, and  HP, IBM in  management of technology and analytic business services

-          Work on projects with Silicon valley firms on these topics

-          Explore the possibility of research support based on course and project performance

 

 

What (background do you need?): Either possess or learn (in the preliminary boot camp for the course) analytics in linear algebra, probability and statistics; a background in machine learning and/or economics is preferred, though not required

What (should you do?): Register and email your CV to me with your background and interests, including project interests.

 

 

Participating Firms (Initial list): Cisco, IBM, Fitme, HP, Visa, Google, Yahoo, Microsoft

 

 

 

Course Modules:

 

•         1. Module 1: Business Processes and functions (marketing, sales, product development and innovation, supply chain management, support services e.g. call centers) in an enterprise ; knowledge processes, services, analytics

 

•         2. Module 2: Data Mining - Finding relationships and associations between different business variables e.g. correlation between prices and sales volumes, factor analysis to improve manufacturing process yields or hotel/airline revenue yields, or stock performance of individual  traders, effectiveness of (web) marketing campaigns, web ad revenue optimization

 

•         Module 3: Knowledge Management, Machine Learning, and Web Search and Mining: Text  and non-numerical Data Mining, Supervised, Semi-supervised, and unsupervised learning, Web search, and information mining and capture, storage, pre-processing, and retrieval for effective enterprise functioning (marketing, support services, products) and Service Center Management

 

•         Module 4: Business Management Data Analytics: Optimization of Marketing portfolios, New Product Development, Financial portfolios , Supply Chain Management, Global Outsourcing

 

•         Module 5: Web economics: Ads, game theory and auctions, Ad Revenue optimization, Recommender systems and Shop Bots

 

 

 

 

Course Description

ISM 206, ISM250 and ISM251 form a three course sequence, and is part of the following ISTM/TIM streams:

-  Information Retrieval and Knowledge Management Sequence, with 245 (Data Mining) and   260(IRKM)

and

- Management of Technology and Services sequence (including 206 (Optimization) and 207 (Stochastic models)

You will learn some of the following skills:

-        Learn why leading firms (and startups) such as as Google, Yahoo, Microsoft, IBM, HP, Fair Isaacs, and Cisco, are focusing on data and business analytics in providing “knowledge management analytics and services” for complex enterprises

-        analyze and synthesize business intelligence platform needs at the algorithmic decision making level, including functions such as marketing and sales through data and text mining, new product development speedup, supply chain and e-business optimization, (and possibly financial engineering and risk management).

-        learn basic analytics of data mining, including statistics, supervised and unsupervised learning approaches, search engine and modern information system retrieval

-        analyze intelligent support systems for marketing decisions, including fundamental methods such as conjoint analysis, together with web search, information retrieval, and data mining approaches, for learning about markets and customer preferences, as well as develop mathematical models for optimizing sales, marketing, and pricing decisions in high tech

-        learn basic of constrained optimization and dynamic programming, including value and policy iteration for finite horizon situations, with applications in supply chain management and e-business, as well as web recommender systems

-        perform financial and decision analysis to manage risk and to develop technologies and products that are profitable, by learning financial engineering ideas in real options and portfolios based on constrained optimization

 

-       In addition, mini projects will be used as a vehicle to better understand and apply the methods..

The course format will be 1/2 lectures, and 1/2 external and internal speakers and seminars including industry personnel, faculty, and students. Significant time will be devoted to project modelling and analysis, and a term (project/research) paper.

The course emphasis will be tuned to the class composition and interest.

 

 

COURSE OUTLINE, SYLLABUS & READINGS   

Week 1:

-       Business and Management Functions (Innovation – R&D/Product Development/ Engineering, Marketing, Finance/Accounting,  Operations/Supply Chain Management,  Organizational Management) for Technology Development and Commercialization

-       Role of data and business analytics and knowledge management in a services economy

-       Distiction between knowledge services and product manufacturing

-       Role of information (“Knowledge”)

-       Arbitrage opportunities fewer? Reputation and communities? Expand Social Network Dynamics, Trust

-       Introduction to Data Mining

-       Differentiation is cruicial (need to give customers some reason to come to you vs. competitor

 

 

Week 2:

- Quantitative market assessment of technology:  Conjoint anlysis, marketing engineering and optimization

- Data Analysis and metrics/goals in data mining: Data Exploration and performance measures

- Knowledge Services and Analaytics and role of Service and Call Centers

 

Week 3

- Constrained Optimization 1 with marketing and product portfolio examples, including pricing

- Constrained  Optimization 2 (Kuhn-Tucker Conditions)  with advanced marketing examples, inlcuding web page layout to maxinmize profits

- Constrained optimization 3: Applied to Product Portfolios and financial portfolios

 

- Data Mining  applications in marketing, sales, credit rating, text/document classification, anomalies etc.), including classification metrics

- Regression in Data Mining 2 – Detailed algorithms (and examples including text mining)

 

Week 4

- Principal components

 - Text analytics

 

Week 5

- Tiered Service Centers for back end service

- Shop bots and search

 

Week 6

- Bayesian Classifiers 1 : Naïve Bayes Classifierse

- Nearest Neighbour Classifiers

 

Week  7

- Stochastic Dynamic Programming (DP) concepts

- DP in Supply Chain Management and E-Business

- Logistic regression in data mining and Decision Trees

 

 

Week 8

- Ad sense and ad optimization

- Game theory market mechanisms

 

Week 9

- Stochastic Dynamic Programming concepts applied to Recommender Systems in Shop Bots
- Learning in services

 

Week 10 (plus extra class)

- Services marketing

- Global delivery models and outsourcing

- Course summary

 

 

Analytic Bootcamp by TA

Weeks 1-5:   

§                                  Review of  Linear Algebra

§                                   Review of statistics

§                                                    Convex sets and functions

§                                   Review of stochastic processes and Markov Chains

§                                                    Basic Net Present Value Concepts

 

Software 

·         Excel and XLMiner add-on

·         Matlab

·         SAS etc. as needed

 

 

Course Grading (May alter to weight project/term/research paper more heavil, if of sufficiently high quality)

 Weekly Homework on fundamental topics, quizzes, Comprehensive Course Project/term paper (including presentation to class)

Homework: 20%

Quizzes and final: 35%

Project/Term paper: 35%

Presentation: 10%

 

 

Textbooks 

Data Mining for Business.Intelligence,  Shmueli, Patel, and Bruce, Resampling Stats, 2006

ISM 209 Course Reader

Secondary Textbooks  (Shared and possible reserves)

1.        Analyzing Multivariate Data: Lattin, Carroll, and Green, Thompson, 2003 (Strongly recommended)

2.        Data Mining Techniques: Berry and Linoff, Wiley, 2nd edition, 2004

3.        Mining  the Web, Soumen Chakrabarti, Morgan Kaufman, 2003 (possibly new 2006 draft version)

4.        Modeling the Internet and the Web, Baldi, Frasconi, Smyth, 2003, Wiley

 

Secondary References (An extensive reference list is being provided for course projects and to help with follow course tracks)

Linear Algebra

-        Linear Algebra, 3rd edition: Strang, Wellesley-Cambridge Press, 2003

-        Matrix Analysis and Applied Linear : Meyer, SIAM, 2000

Statistics

-        Analyzing Multivariate Data: Lattin, Carroll, and Green, Thompson, 2003 (Strongly recommended)

-        Statistical Models: Freedman, Cambridge Press, 2005

-        Introduction to Probability and Statistics: Ross, Wiley, 1987

Mathematical Programming, Stochastic Models/Processes, and  StochasticOptimization

-        Nonlinear Programming, Mokhtar Bazaraa, and CM Shetty, Wiley, 1979

-        Nonlinear Programming: Avriel, Dover, 1976/2003

-        Mathematical Optimization and Economic Theory: Intriligator, SIAM, 1971/2002

-        Stochastic Processes, Sheldon Ross, Academic Press, 1993

-        Queueing Systems, Gross and Harris, 1993

-        Dynamic Programming and Optimal Control, Vols 1- 2, 2nd ed, Dimitri Bertsekas,  Athena Scientific, 2000 & 2002

Supply Chain Management and E-Business

-         E-Business and Supply Chain Networks, Simchi-Levi et.al, Kluwer, 2003

-        Supply Chain Management: Tayur and Magazine, Kluwer, 1998

-        Supply Chain Management and E-Business: Management Science Special Issue (and Interfaces) 2003

-        E-Business Management: Ed. By Shaw, Kluwer, 2003 (Strongly recommended)

Data and Text Mining

-        Pattern Classification, 2nd Edition: Duda, Hart, and Stork, Wiley,  2001 (Strongly recommended)

-        The Elements of Statistical Learning: Hastie, Tibshirani, and Friedman, Springer, 2001

-        Data Mining: Witten and Frank, Elsevier/Morgan Kaufmann, 2005

-        The Search: Battelle, Portfolio, 2005

-        Principles of Data Mining, David Hand, Heikki Mannila, Padhraic Smyth, Prentice Hall, 2001

-        Data Mining Techniques: Berry and Linoff, Wiley, 1997

-        Data Mining: Han and Kamber, Morgan kaufmann, 2001

-        Data Mining: modeling Data for marketing, Risk, and CRM: Rudd, Wiley, 2001/2003

-        Understanding Search Engines, Michael Berry and Murray Browne, SIAM, 1999

-        Modern Information Retrieval, Ricardo Baez-Yates, Berthier Ribeiro-Neto, Addison-Wesley, 1999.

AI and Machine Learning

-        Pattern Recognition and Mchine Learning: Bishop, Springer, 2006

-        Introduction to Machine Learning, Alpaydin, MIT Press, 2004

-        Artificial Intelligence,, A Modern Approach,2nd edition: Russell and Norvig, Prentice Hall, 2002

-        Introduction to Knowledge Systems, Mark Stefik

New Product Development

-        Management Science: Special Issue on New Product development, 2001

-        Setting the Pace in New Product Development: McGrath, Elsevier, 1996

-        Product Leadership: Cooper, Basic Books, 2005

-        Developing Products in Half the Time, 2nd edition: Smith and Reinertsen,  1998

-        The Balanced Scorecard: Kaplan and Norton, HBS, 1996

Entrepreneurship

Engineering Your Startup: Baird, Professional Publications, Inc., 1999 (Strongly recommended)

Marketing and/or Search

-        Search Engine Marketing, Inc.: Moran and Hunt, IBM Press, 2006

-        Marketing Research, 4th edition: Malhotra, Prentice Hall, 2004

-        Marketing Management, Phillip Kotler, Prenctice Hall, 2002.

-        Marketing Engineering, Gary Lilien and Arvind Rangaswamy, Prentice Hall, 2003

-        Marketing Models, Gary Lilien, Philip Kotler, Sridhar Moorthy, Prentice Hall, 1992

-        Intelligent Support Systems for Marketing Decisions, Nikolaos Matsatsinis and Yannis Siskos, Kluwer, 2003

Finance and Financial Engineering

-        Investment Science: Luenberger, Oxford University Press, 1998

-        Principles of Corporate Finance, 7th Edition: Brealey and Myers, McGraw-Hill, 2003

-        Corporate Finance, 7th edition: Ross, Westerfiled, and Jaffe, McGraw-Hill, 2005

-        Real Options: Trigeorgis, MIT Press, 1996

-        Real Options: Amram and Kulatilaka, HBS, 1999

-        Financial Engineering by Stan Pliska, 2001

-        Theory of Financial Decision Making: Ingersoll, Rowman & Littlefield, 1987

-        Financial Modeling in Excel: Benninga, MIT Press, 2001

 

Strategic Management

Strategic Management: Saloner, Shepard, and Podolny, Wiley, 2001

 

Prerequisites:

Wile no formal prerequisites are required, to provide for diverse student backgrounds from AM/BME/CS/CE/Econ/EE//TIM, it is assumed that students have undergraduate preparation equivalent to the probability and statistics level of CE 107, and possibly some exposure to linear algebra. Instructor approval based on mathematical maturity is a possibility. A boot camp in probability, statistics, and linear algebra is planned, to bring all the students to a common level.

Course Context

This course is intended to be the one in a series of courses in the new Technology  and Information Management Program, covering Knowledge Analytics and Management in Business and Services. The motivation for these courses is to teach students the theory and practice of the technology development and management, through the use of information system based decision making.

This sequence of core courses will  form the foundation on which other graduate courses in TIM will be built.