MCS314(Semester I 2007) Mining of Streaming Data

Being an advanced course in Data Mining, the pre-requisite is MCS104. Other pre-requisite is a course on Algorithms. Strong programming skills and dexterity with complex data structures will be of advantage. Book by Charu Aggarwal will form the basic reading material. Following papers will be discussed.

1. Mining Data Streams: A Review

2. Mining HighSpeed Data Streams

3. Mining Time Changing Data Streams

Tentative schedule

Week 1	Overview of Data Mining techniques Introduction to mining of Data Streams (DS)
Week 2 -5	Clustering of Data streams
Week 6	Minor 1
Week 7- 9	Mining frequent item-sets from DS
Week 10	Minor II
Week 12 - 14	Change Detection and Change  Modeling in DS

Internal Assessment

2 Minors as per the schedule displayed on the department notice board (25 marks)

Programming Assignment (15+5 marks): Implement one of the two algorithms i) Clustream   ii) DenStream. Write a detailed report on the data structures used, other implementation issues (Submit by 12 Nov.  2007)

Presentation of experimental  analysis of a research paper related to data stream mining   (10 marks) (During 4rd week Oct. 2007)

Take Home Assignments as and when given in class (10 marks)

Assignment I : (Submit by 5 Nov.) Write a formal algorithm for Clustream and Denstream, and work out the complexity of the online and offline components of the two algorithms.

Assignment II: (Submit by 12 Nov.) Work out the time and space complexity of the change modeling algorithm proposed in MONIC