The document summarizes information about a group project involving data stream clustering. It lists the group members and then discusses key concepts related to data stream clustering like requirements for algorithms, common algorithm types and steps, prototypes and windows. It also touches on outliers and applications of clustering.

1. MEMBERS:
Dheeraj Pachauri(1809113042)
Himanshu Bharti(1809113052)
Shahnawaz Khan(1900910139007)
Abhay Kumar Mishra(1900910139001)

2.  Clustering
 Data Stream
 Stream Clustering
 Requirements for clustering algorithms
 Stream clustering steps & algorithms
 Prototype array
 Window models
 Outliers & its detection
 Applications of clustering

3.  Method of identifying similar groups of data in a data
set.
 Entities in each group are comparatively more similar
to entities of that group than those of other group.
 Some methods include K-means, K-mediods, DB-
SCAN etc.

4.  STREAM: Data that arrives continuously such as Google
queries, telephone records, multimedia data, financial
transactions etc.
 Not feasible to store in a database & data can be lost if not
processed immediately
 DATA STREAM: Continuous, massive, unbounded
sequences of data objects that are continuously generated at
a rapid rate.
 The problem of data stream clustering is defined as:
Input: a sequence of n points in metric space & an
integer k.
Output: k centers in the set of the n points so as to
minimize the sum of distances from data points to their
closest cluster centers.

5.  ONLINE PHASE
 Summarize the data into memory-efficient data
structures
 OFFLINE PHASE
 Use a clustering algorithm to find the data
partition

6.  Provide timely results by performing fast &
incremental processing of data objects
 Rapidly adapt to changing dynamics of the data,
which means algorithm should detect when new
clusters may appear, or others disappear
 Scale to the number of objects that are
continuously arriving
 Provide a compact model representation
 Rapidly detect the presence of outliers & act
accordingly
 High dimensionality, interpretability & usability
 Deals with different data types. Ex- XML trees,
DNA sequences, GPS information etc.

7.  ALGORITHM STEPS:
 Data Abstraction: Summarize the data into
memory-efficient data structures
 Clustering phase: Use a clustering algorithm to
find the data partition

8. There are five main classes:
 HIERARCHICAL BASED ALGORITHMS: It
uses the dendrogram data structure which is
binary tree based. Useful to summarize &
visualize the data.
 Examples are BIRCH, CHAMELEON, ODAC,
E-Stream & HUE-Stream.

9.  It splits the data instances into a predefined
number of clusters based on similarity to the
cluster centroids.
 Examples are Clustream, HPStream,
SWClustering, StreamKM++ & CLARA.

10.  It uses multi-resolution grid data structure.
 The workspace is divided into a number of
cells, in a grid structure, and each instance is
assigned to a cell
 Grid cells are then clustered.
 Examples include GCHDS, GSCDS, DGClust,
CLIQUE, WaveCluster & STING.

11.  It keeps summary of input data in large
number of micro clusters.
 Micro cluster is a set of data instances that are
very close to each other.
 Synopsis is kept with a feature vector. Then,
these micro clusters are merged & formed final
clusters.
 Examples are DBSCAN, LDBSCAN, DSCLU,
SOStream & MR-Stream

12.  It finds the data distribution model that fit best
to the input data.
 Attempt to optimize the fit between the data &
some mathematical model.
 Adopts statistical & AI approach
 Examples are COBWEB, CluDistream & SWEM

13.  Some data stream clustering algorithms usea
simplified summarization structure called
prototype array.
 Array of protoypes that summarizes the data
partition.
 It’s used to summarize the stream to divide the
data stream into chunks of size m.

14.  In most data stream scenarios, more recent
information from the stream can reflect the
emerging of new trends or changes on the data
distribution.
 This information can be used to explain the
evolution of the process under observation.
 Moving window techniques have been
proposed to partially address this problem.

15.  Only the most recent information from the data stream are stored
in a data structure whose size can be variable or fixed.
 This is usually a first in, first out(FIFO) structure which considers
the objects from the current period of time upto a certain period in
the past.
 The organization & manipulation of objects are based on the
principle of queue processing.

16.  Considers the most recent information by associating
weights to objects from the data stream.
 More recent objects receive higher weight than older
objects & the weights of the objects decrease with time.
 The weight of the objects exponentially decays from
black to white.
 Adopted in density based clustering algorithms.

17.  Last in the row
 It considers the data in the data stream from
the beginning until now.

18.  The coreset tree structure is responsible for
reducing 2m objects to m objects. The
construction of this structure is defined as
follows:
 First, the tree has only the root node v, which
contains all the 2m objects in Ev. The prototype
of the root node Xpv is chosen randomly from
Ev & Nv=|Ev|=2m. Afterwards, two child
nodes for v are created as v1 & v2.
 To create these nodes, the object that is farthest
away from the prototype object is selected.

19.  OUTLIERS: The set of objects are considerably dissimilar from
the remainder of the data.
 PROBLEM: Find top n outlier points
 APPLICATIONS:
 Credit card fraud detection
 Telecom fraud detection
 Customer segmentation
 Medical analysis

20.  Besides the requirements of being incremental
& fast, data stream clustering algorithms
should also be able to properly handle outliers
through the stream.
 These are objects that deviate from the general
behaviour of a data model & occur due to
different causes, such as problems in data
collection, storage & transmission errors,
fraudulent activities or changes in the
behaviour of the system.

21.  Pattern recognition
 Spatial data analysis
 Image processing
 Economic Science(especially market research)
 WWW

22.  Internet
 Data Mining & Analysis by MJ Zaki
 Websites(dimacs.rutgers.edu &
dsc.soic.indiana.edu)
 Class notes