CSCE 970 Lecture Transcriptions

This semester each student in CSCE 970 will transcribe some of the lectures for the benefit of the student and the class. The number of lectures each student will be approximately 2.5, depending on when specific topics are fully covered. The transcription process requires the student to take copious notes of the assigned lectures and write a brief but detailed summary of it. The student scribe must submit an electronic copy of the summary (formatted similarly to the following template and example) to me no later than two weeks after the lecture. The submission should include (1) a PostScript version, (2) an ASCII version, and (3) the original source from whatever document processing package used to generate it. Though I strongly prefer LaTeX for the source (use the following LaTeX template to aid with formatting and the following tutorials to learn about the package), the source from other document processors is acceptable. If you use LaTeX, then you do not have to submit an ASCII version (i.e. only (1) and (3), not (2)). The PostScript versions will be posted on this web page so the other students can refer to it.

I will assign specific lectures to students only a few weeks in advance, but if there is a particular topic that interests you, you may send me e-mail to request priority on lectures on that topic.

o Example.

o PostScript Template.

o LaTeX Template.

o LaTeX Tutorials

The Transcriptions
o Bayesian-Based Classifiers
o Estimating Unknown pdf's
o Perception Algorithm, Winnow, Exponentiated Gradient
o Non-Linearly Separable Classes and Nonlinear Classifiers
o The Backpropagation Algorithm and its Variations
o Sizing the Network, Generalized Linear Classifiers, Decision Trees
o Feature Selection
o Class Separability Measures
o Feature Subset Selection
o System Evaluation: Estimating classification error, boosting
o Introduction to Clustering: Applications, cluster types, feature types, definitions of proximity measures
o Introduction to Clustering (cont'd): Examples of proximity measures, categories of algorithms
o Sequential Clustering Algorithms
o Hierarchical Clustering Algorithms
Return to the CSCE 970 Home Page