CS 568 Data Mining
Notes for Lecture 1
Desh Raj (130101018), Dept. of CSE
August 16, 2016
1
Introduction
In this lecture, we will first look at why data mining is important. When we have
established the necessity for this course, we will proceed to an int
Lecture Date: 1 November 2016
Name: Mrinal Tak
Roll No . 130101049
BTech CSE
Hierarchical Clustering
We create a sequence of partitions ranging from each point in its own cluster to one cluster
containing all points.
There are two approaches in hierarchal
Data Mining
FALL SEMESTER 2016
INSTRUCTOR: Prof. Amit Awekar
Jitendra Choudhary (130101017)
29 August 2016
Frequent Substring Mining
Frequent patterns are itemsets, subsequences, or substructures that appear in a data set
with frequency no less than a use
SEQUENTIAL PATTERN MINING
1 September 2016
130101025
Till now we have tried to mine the frequent item-sets
based on their occurance in a single transaction. We now
look for patterns across transactions.
Lets consider an example. Suppose we have records
ab
06 Sept 2016
Divya D Kulkarni
166101005
Maximal Clique Enumeration
Identifying all maximal cliques in a simple graph.
Consider the Graph as below. The Goal is to find out all the maximal cliques from the Graph.
By Manual Inspection, we find the maximal cl
CS 568 Data Mining
Lecture Note, November 2, 2016
Rahul Kumar Gond (130101061), Dept. of CSE
Continuation Previous class
Bottom up hierarchical clustering: In bottom up or agglomerative hierarchical clustering, we begin with each of the n points in a sepa
Data Mining
Lecture Notes
Topic : Frequent Substring Mining using Suffix Trees
Akhil Polamarasetty
8-29-2016
Topics discussed so far:
We have looked at various Frequency item set mining algorithms.
Aprioris Algorithm, Prefix and Suffix based approaches
Frequent Sequence Mining
SPADE (Sequential Pattern Discovery using Equivalence classes)
DataMining(CS568)
LectureNotes(Dated:30/08/2016)
Instructor:Prof.AmitAwekar
Name:RaviKumar
RollNo.:130101064
SPADE:
The sequence mining task is to discover a set of at
Lecture Notes [23rd Aug]
Instructer: Dr Amit Awekar
Ayush Kumar cfw_130101011
In the previous class we were discussing about Frequent itemset generation in Apriori Algorithm by using the fact that If an itemset is frequent
then all of its subsets must al
Scribe
Varun Raj
25th August 2016
1
Revision
Problems with Apriori Method The problem with Apriori algorithm is support
counting, every time we need to count support for a new data item we need to
go through the whole record. Assuming that data is large e
Expectation Maximization Algorithm for Clustering
Eeshani Mondal 130101021
3rd November, 2016
Expectation Maximization
EM is based on the following core ideas :
That there exists an analytic model for the data and that we know the functional
form of the m
Frequent Graph Mining: DATA MINING
FALL SEMESTER 2016
INSTRUCTOR: Prof. Amit Awekar
Reference :
http:/www.dataminingbook.info/uploads/Main/BookPathUploads/book-20160121.pdf
09 September 2016
DFS Code: Edge Ordering
Let eij = (vi ,vj ) and exy = (vx ,vy )
Lecture Notes: 06 September 2016
Sanjukta Dutta 166101004
Maximal Clique Enumeration
The maximal clique enumeration (MCE) problem is the problem of enumerating all maximal
cliques in a graph. Informally, a graph or undirected graph G(V, E) is a set of ver
Maximal Frequent Itemsets
A maximal frequent itemset is, frquent itemset, such that, none of its supersets are frequent.
Consider the following transactions,
1. ac
2. bd
3. abc
4. b
5. cd
6. abd
7. ab
8. c
9. cd
10. b
11. abc
Threshold = 2
abcd
cfw_null
a
Lecture Notes: - 26-AUG-2016
Anirudh Agnihotry
130101007
Storing of all the frequent item sets require a large amount of space. So we want some
condensed representation for storing the frequent item sets. This condensed representation
may or may not lead
LECTURE NOTES
9TH SEPT 16
DEEPIKA BISHNOI
166151009
FINDING OUT THE MINIMUM DFS CODE:
1
DFS CODE:
1,2
2,3
DFS CODE:
1,2
1,3
1
2
2
3
F
2,3
F
1,3
eij
exy
j=y
i>x
Therefore eij < exy
Therefore 2,3 < 1,3
3
Which one of these two is correct DFS code
for th
Lecture notes:- 30- AUG- 2016
Sudip Suprakash Pati
164101054
Frequent Sequence Mining
In frequent sequence mining, we have to consider all the permutations of the
symbols as the candidate for frequent sequences instead of only combinations
of items as in
Data Mining Notes :
Name Ajinkya Sanjay Mankar
Roll No- 164101059
Mtech I (CSE)
Eclat Algorithm
The Eclat algorithm leverages the tidsets directly for support computation. The
basic idea is that the support of a candidate itemset can be computed by inter
CS 523 Advanced Computer Architecture
Lecture 5 & 6
Introduction to MIPS Pipeline
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
MIPS Pipeline Outline
Pipelined Vs Unpipelined De
CS 523 Advanced Computer Architecture
Lecture 14
Superscalar Processors with Multiple Issue
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
Multiple Issue and Static Scheduling
Ba
CS 523 Advanced Computer Architecture
Lecture 7
Branch Prediction Techniques
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
Four Branch Hazard Alternatives
#1: Stall until branch
CS 523 Advanced Computer Architecture
Lecture 12
Dynamic Scheduling by Tomasulos Algorithm
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
Tomasulos Algorithmv
Steps in Tomasulos A
CS 523 Advanced Computer Architecture
Lecture 9
MIPS Pipeline for Multi-cycle Operations
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
Multi-cycle Operations
Some operations req
CS 523 Advanced Computer Architecture
Lecture 2
Review of basic computer organization
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
Execution Cycle
Instruction
Fetch
Obtain instr
CS 523 Advanced Computer Architecture
Lecture 1
Introduction and Course Overview
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
About Me
I hail from a small village in central Ker
CS 523 Advanced Computer Architecture
Lecture 3
Performance Measurement & ISA
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
Measuring Performance
When can we say one computer /
CS 523 Advanced Computer Architecture
Lecture 11
Branch Prediction & Dynamic Scheduling
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
Branch Prediction Schemes
Basic 2-bit predic
CS 523 Advanced Computer Architecture
Lecture 13
Hardware Based Speculation
John Jose
Assistant Professor
Department of Computer Science & Engineering
Indian Institute of Technology Guwahati, Assam.
Hardware-Based Speculation
How to overcome control haza