Edo Liberty Ph.D

Research Scientist
Yahoo! Research
Haifa, Isreal
edo.liberty@ymail.com
עידו ליברטי
homepage

Class description:

Data Mining is concerned with efficiently extracting statistics, patterns, structures, or meanings from raw data. This task becomes hard when the amount of data is large, which is often the case in modern data sets. This course will survey modern algorithms, concepts, and data structures common to data mining in large data sets. We will try to cover, among other topics: data sampling, finding causal relations and frequent item sets, counting in data streams, ranking and sorting, approximating large matrix operations, dimensionality reduction and efficient searching in high dimensions. We will also discuss modern cluster architectures and computational models.

I recommend that students be familiar with probability theory, basic combinatorics, linear algebra, basic complexity theory, and traditional data structures, at least on at introductory level. The class will attempt to be self contained nonetheless.

Some details...

Undergraduate students:

• Final exam (50% of the grade).
• Home assignments (50% of the grade).
• A project (for extra credit) is optional for students who are interested.

• No final exam.
• Home assignments (50% of the grade).
• Final project (50% of the grade). Should require roughly a week's worth of work, comparable to learning for and taking the final exam. These project can contain both theoretical and experimental elements.

Class notes