09-03-2012, 02:18 PM
Student Performance usingK-means algorithm
1st review.ppt (Size: 1.43 MB / Downloads: 69)
INTRODUCTION
The problem is to maintain entire students information and evaluating the performance based on their attendance & academics.
Also the students data can be clustered to retrieve the information under various categories.
The internal and external marks of every subject and attendance is taken and they are clustered and performance is evaluated.
Techiniques of DataMining :-
Association:- It finds interesting correlation among a
large set of data items.
Classification:- It is used to place data elements into
related groups with prior
knowledge of group definitions.
Clustering:- It is used to place data elements into
related groups without prior
knowledge of group definitions
Prediction:- It is form of data analysis used to extract
models describing important data
classes to predict future data trends
State of Art
Types of clustering methods
Partitioning methods:-
It is the method in which “n” data objects
can be classified into “k” clusters (groups). [k< n].
ex:- K-means, K-mediods.
Hierarchical methods:-
1)Agglomerative (“bottom – up”)
2)Divisive(“top – down”)
Agglomerative:- The approach starts with each object forming a separate group later it successively groups the objects close to one another.
Divisive:- The approach starts with considering all the objects into the same cluster later It splits into smaller clusters.
ex:- BIRCH ,CURE
Proposed Method
K-Means Algorithm:-
The K-means algorithm takes the input parameter ,k,and partitions a set of n objects into K clusters so that the resulting intera cluster similarity is high whereas the inter cluster similarity is low .
The algorithm proceeds as follows: first, it randomly selects K of the object which each represent a cluster mean of center .for each of the remaining objects ,an object is assigned to the cluster to which it is the most similar.based on the distance between the object and the cluster mean.