This document summarizes a study analyzing students' academic performance data. The study collected internal and external marks for 45 students over 5 semesters. It cleaned the data, transforming the marks into sums, and used k-means clustering to group students into 4 categories (excellent, good, fair, poor) for each semester based on their internal and external marks. The analysis found the clusters followed the same performance pattern each semester, with students scoring higher internally also scoring higher externally, indicating a direct relationship between internal and external marks. The study concluded a student's university exam performance can generally be predicted from their internal marks.

1. Students Academic
Performance
Knowledge Discovery from Data

2. Introduction..
 Our project aim is to find students academic performance
and find out whether there is any general pattern in their
marks and performance.
 So here ,We are analyzing both internal and external
marks of a student.
 We did the following KDD preprocessing steps to mine
our data.

3. Learning the application domain
 Learning the application domain is the first step in KDD
process .
 Need to have a clear understanding about the application
domain and our objectives.
 The institution considered for mining is MCA batch of Rajagiri
College of Social Sciences.
 We collected all previous year academic record from the
department of computer science

4. Create a target data set:
data selection
 We selected 2007-2010 batch marks for analysing the
pattern.
 There were around 45 records(45 students).
 Both the internal and external marks of each student were
selected, in order to find out the performance pattern.

5. Internal & External Dataset

6. Data cleaning & preprocessing
 Data cleaning is the step where noise and irrelevant data are
removed from the large data set.
 This is a very important pre-processing step because our
outcome would be dependent on the quality of selected data.
 Remove duplicate records, enter logically correct values for
missing records(absent students), remove unnecessary data
fields and standardize data format.

7.  There was no much duplicate data or unnecessary data in the
collected record . The dataset was partially cleaned.
 Student internal mark and external mark were stored in
different records.
 By applying data integration these records were integrated
into one record.
 The new dataset consist of internal mark details and external
mark details of each student in one record.

9. Data reduction & transformation
 Data is transformed into appropriate form for making it ready for
data mining step.
 The dataset contains marks of 5 theory paper and 2 lab paper of
all 5 semesters.
 These marks are transformed into sum of internal marks and sum
of external marks of each student for the easiness of analysing
the pattern.

11. Cluster Analysis
 The data mining technique we used here is clustering.
 A cluster is a collection of data objects that are similar to
one another within same cluster and are dissimilar to
objects in other cluster.
 We first partitioned the set of data into groups based on
data similarity and then assign labels
Choosing functions of data mining

12. K-MEANS Partitioning
 The K-means algorithm takes input parameter k and
partitions the set of n objects into k clusters.
 Here we selected no: of cluster as 4
 Objects are distributed to a cluster based on cluster
center to which it is nearest.
 For each semester we found out the clusters separately
and labeled them as students Excellent, Good, Fair and
Poor
Choosing mining algorithms

13. The Tool used for pattern evaluation is ORANGE

14. Orange Cluster Analysis

15. No of cluster selected is 4

16. Semester 1
poor
Fair
Good
Excellent

17. Semester 2

18. Semester 3

19. Semester 4

20. Semester 5

21. Centroid Analysis

22. Semester 1

23. Semester 2

24. Semester 3

25. Semester 4

26. Semester 5

27. Combined Centroid Analysis

28. Data mining search for patterns of
interest
 From the mining process we found that “All the 5 semester
clusters followed the same pattern of performance”.
 A student with high internal mark has higher external
marks and a student with less internal marks has less
external marks.
 There is a direct relation between the internal and the
external marks.
 At some case this evaluation is not valid, cases like
 Being absent for internal exam and scoring high marks for
the externals (vice versa)

29. CONCLUSION
 A students performance in his university exam can be
predicted with the help of his internal marks. There is
a direct relation between the internal and the external
marks.
 A student with low internals will get low marks for
externals too

30. Use of discovered knowledge
representation

31. Thank You