Lecture Details and Reading Material

Collaborative Adaptive Sensing of the Atmosphere
Prof. Jim Kurose

Lectures

  • Overview of a data-driven sense-and-response system (1 hour)
  • Following the data: from sensor signals to sirens (1 hour)

This lecture will cover topics such as data acquisition, averaging, compression, storage, detection, tracking, and display.

  • Networked radar systems (3 hours)

This lecture will cover topics such as networking architecture of the radar, congestion control and routing, and networking off-the-grid-radars.

  • Computational infrastructure (1 hour)

This lecture will review the meteorological command and control infrastructure, its pub/sub as well as blackboard systems.

  • Working with end-users (1 hour)
  • Open issues (1 hour)

Reading material

http://gaia.cs.umass.edu/kurose/tcs/tcs_kurose_reading_v1.htm

Algorithms for Analyzing Massive Streams
Professor S. Muthukrishnan

Lectures

  • Overview of massive data applications and requirements (1 hour)
  • Data Stream Models, Sublinear space/time Algorithms, and Applications to IP network traffic analysis. (4 hours)
  • Compressed Sensing and applications to signal processing (1 hour)
  • Massive distributed streams and Web traffic analysis.  (1 hour)
  • Open Problems (1 Hour)

Reading material

S. Muthukrishnan’s web site (with several pointers to data stream algorithms):
http://www.cs.rutgers.edu/~muthu/

Reference: Data Streams and Algorithms, S. Muthukrishnan, Book details at
http://www.nowpublishers.com/tcs/

Compressed Sensing and applications to signal processing
http://www.cs.rutgers.edu/~muthu/ncs.pdf

Massive distributed streams and Web traffic analysis. Reference:Mapreduce paper at
http://labs.google.com/papers/mapreduce-osdi04.pdf

Trends in Databases—Mining, Exploratory Analysis, and the Web
Raghu Ramakrishnan

Lectures

  • DBMS Support for Complex Data Analysis (2 hours)
    • OLAP, Warehousing, View Materialization
    • Sorted relations, streaming data, continuous queries
  • Exploratory Mining (2 hours)
    • Overview of Data Mining, SQL Support
    • Combining OLAP with Mining
  • Web Data Management (2 hours)
    • Searching the Web
    •  Managing Data Extraction and Integration

Reading material

  • R. Ramakrishnan and J.G. Gehrke: Database Management Systems, McGraw-Hill, 3rd ed. 
  • J. Melton, A. Simon: Understanding SQL:1999—Object-Relational and Other Advanced Features, Morgan Kaufmann, Chapter 7 (OLAP).
  • A. Netz, S. Chaudhuri, U. Fayyad, J. Bernhardt: Integrating Data Mining with SQL Databases: OLE DB for Data Mining, ICDE 2001.
  • S. Brin, L. Page: The Anatomy of a Large-Scale Hypertextual Web Search Engine, WWW 1998.
  • A. Doan, R. Ramakrishnan, S. Vaithyanathan: Managing Information Extraction, Tutorial, SIGMOD 2006
  • Ramakrishnan, B-C. Chen: Exploratory Mining in Cube Space, Journal of Data Mining and Knowledge Discovery, 2007.
  • Amir Netz, Surajit Chaudhuri, Jeff Bernhardt, Usama Fayyad. Integration of Data Mining and Relational Databases. Proceedings of the 26th International Conference on Very Large Databases, Cairo, Egypt, 2000
  • R. Ramakrishnan and Bee-Chung Chen. Exploratory Mining in Cube Space.

An Introduction to Support Vector Machines and other Kernel Methods
Dr. Alex Smola 

This course will give an introduction to kernel methods.  In particular, it will include an overview of Support Vector Classification and Regression, Novelty Detection, and Quantile Regression. Several kernels will be discussed and how they can be used in a number of algorithms, ranging from Kernel PCA to Perceptrons, will be demonstrated. It will conclude with an overview of Structured Estimation, in particular as it can be used for Named Entity Recognition and ranking of Web pages for an Internet search engine.

Reading material

Learning with Kernels
http://www.learning-with-kernels.org

The first chapter of Learning with Kernels is available for download at:
http://www.learning-with-kernels.org/sections

A tutorial on Support Vector Classification 
http://research.microsoft.com/~cburges/papers/SVMTutorial.pdf

A tutorial on Support Vector Regression
http://sml.nicta.com.au/Publications/homepublications/publications/papers/2004/SmoSch04.pdf

A recent paper that  has been submitted to the annals of statistics (A Review of Kernel Methods in Machine Learning)
annals_re-submission.pdf

Some software can be found at
http://elefant.developer.nicta.com.au