Research scholars evaluation based on guides view

IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 02 Issue: 10 | Oct-2013, Available @ http://www.ijret.org 380
RESEARCH SCHOLARS EVALUATION BASED ON GUIDES VIEW
USING ID3
Sathiyaraj.R1
, Sujatha.V2
1, 2
Assistant Professor, Department of CSE, MITS, SVCET, sathiyarajr@mits.ac.in, suji4068@gmail.com
Abstract
Research Scholars finds many problems in their Research and Development activities for the completion of their research work in
universities. This paper gives a proficient way for analyzing the performance of Research Scholar based on guides and experts
feedback. A dataset is formed using this information. The outcome class attribute will be in view of guides about the scholars. We
apply decision tree algorithm ID3 on this dataset to construct the decision tree. Then the scholars can enter the testing data that has
comprised with attribute values to get the view of guides for that testing dataset. Guidelines to the scholar can be provided by
considering this constructed tree to improve their outcomes.
----------------------------------------------------------------------***-----------------------------------------------------------------------
1. INTRODUCTION
Data mining an interdisciplinary subfield of computer science,
is the computational process of discovering patterns in large
data sets involving methods at the intersection of artificial
intelligence, machine learning, statistics, and database
systems. The overall goal of the data mining process is to
extract information from a data set and transform it into an
understandable structure for further use. The actual data
mining task is the automatic or semi-automatic analysis of
large quantities of data to extract previously unknown
interesting patterns such as groups of data records (cluster
analysis), unusual records (anomaly detection) and
dependencies (association rule mining). This usually involves
using database techniques such as spatial indices. These
patterns can then be seen as a kind of summary of the input
data, and may be used in further analysis or, for example,
in machine learning and analytics. Data Mining can be used to
solve many real time problems. Decision tree is an efficient
method that can be used in classification of data. A decision
tree is a decision support tool that uses a tree-like graph or
model of decisions and their possible consequences, including
chance event outcomes, resource costs, and utility. In this
paper, we use decision tree algorithm ID3 for analyzing
feedback given by guides. The training dataset consists of
attributes such as Research proposal, Qualification,
Experience, Way of Problem solving, Knowledge level,
Interaction with guide, Journals published, Implementation of
algorithm, Relating with real-life applications, Assessment,
Subject knowledge, Punctual and Nature. The outcomes in the
training dataset are specified with values like Excellent, Good,
Poor and Average. The ID3 Algorithm can be applied on this
training dataset to form a decision tree with view of guide as a
leaf node. Whenever any research scholars provide testing
data consisting of attribute values to the formed tree. Also, we
can suggest the possible area where he/she has scope for
improvement. This will help the scholar for self-evaluation
and improvement where they lag.
The Next section describes about the decision tree algorithm
and also defines entropy and gain ratio which are necessary
concepts for constructing decision tree using ID3 and the next
section by describing the problem statement and how we can
analyze the dataset and evaluate the problem by using ID3
algorithm; finally, the conclusions and future works are
outlined.
2. ID 3 ALGORITHM
A decision tree is a tree in which each branch node represents
a choice between a number of alternatives, and each leaf node
represents a decision. Decision tree are commonly used for
gaining information for the purpose of decision -making.
Decision tree starts with a root node on which it is for users to
take actions. From this node, users split each node recursively
according to decision tree learning algorithm. The final result
is a decision tree in which each branch represents a possible
scenario of decision and its outcome.
Decision tree learning is a method for approximating discrete-
valued target functions, in which the learned function is
represented by a decision tree.
ID3 is a simple decision learning algorithm developed by J.
Ross Quinlan (1986) at the University of Sydney. ID3 is based
off the Concept Learning System (CLS) algorithm. The basic
CLS algorithm over a set of training instances C:
Step 1: If all instances in C are positive, then create YES node
and halt.
If all instances in C are negative, create a NO node and halt.

__________________________________________________________________________________________
Otherwise select a feature, F with values v1, ..., vn and create
a decision node.
Step 2: Partition the training instances in C into subsets C1,
C2, ..., Cn according to the values of V.
Step 3: apply the algorithm recursively to each of the sets Ci.
ID3 constructs decision tree by employing a top-down, greedy
search through the given sets of training data to test each
attribute at every node. It uses statistical property call
information gain to select which attribute to test at each node
in the tree. Information gain measures how well a given
attribute separates the training examples according to their
target classification. The algorithm uses a greedy search, that
is, it picks the best attribute and never looks back to reconsider
earlier choices.
2.1. Entropy
Entropy is a measure of the uncertainty in a random variable.
Entropy is typically measured in bits, nats, or bans. It is a
measure in the information theory, which characterizes the
impurity of an arbitrary collection of examples. If the target
attribute takes on c different values, then the entropy S relative
to this c-wise classification. Entropy is formally defined as
follows: If a data set S contains examples from m classes,
then the Entropy(S) is defined as following:
Where Pj is the probability of class j in S
Given a database state, D, Entropy (D) finds the amount of
order in that state. When that state is split into s new states S =
{D1, D2,…, Ds}, we can again look at the entropy of those
states. Each step in ID3 chooses the state that orders spitting
the most. A database state is completely ordered if all tuples in
it are in the same class.
2.2. Information Gain
ID3 chooses the splitting attribute with the highest gain in
information, where gain is defined as the difference between
how much information is needed to make a correct
classification before the split versus how much information is
needed after the split. Certainly, the split should reduce the
information needed by the largest amount. This is calculated
by determining the difference between the entropies of the
original dataset and the weighted sum of the entropies from
each of the subdivided datasets. The entropies of the split
datasets are weighted by the fraction of dataset being placed in
that division. The ID3 algorithm calculates the Information
Gain of a particular split by the following formula:
If attribute A is used to partition the data set S,
Where, v represents any possible values of attribute A;
Sv is the subset of S for which attribute A has value v;
|Sv| is the number of elements in Sv;
|S| is the number of elements in S.
ID3 Algorithm for Decision Tree can be given as
ID3 (Examples, Target_Attribute, Attributes)
1. Create a root node for the tree
2. IF all examples are positive, Return the single-node
tree Root, with label = +
3. If all examples are negative, Return the single-node
tree Root, with label = -
4. If number of predicting attributes is empty, then Return
the single node tree Root, with label = most common
value of the target attribute in the examples
5. Otherwise Begin
5.1 A The Attribute that best classifies examples
5.2 Decision Tree attribute for Root A
5.3 For each positive value, vi, of A,
5.3.1 Add a new tree branch below Root,
corresponding to the test A = vi
5.3.2 Let Examples (vi), be the subset of examples that
have the value vi for A
5.3.3 If Examples (vi) is empty
Then below this new branch add a leaf node
with label = most common target value in
the examples
Else below this new branch add the subtree
ID3 (Examples(vi), Target_Attribute,
Attributes – {A})
6. End
7. Return Root
The ID3 algorithm works by recursively applying the splitting
procedure to each of the subsets produced until “pure” nodes
are found—a pure node contains elements of only one class—
or until there are no attributes left to consider.

__________________________________________________________________________________________
3. DETERMINING IN-GENERAL-VIEW OF
GUIDES ABOUT THE SCHOLARS
The problem that we are considering here is to determine the
in-general-view of guides about the scholar. Based on the
outcome value, we can suggest ways for scholars to improve.
To achieve our goal, we use ID3 algorithm that is described in
the previous section. First, we can have the training dataset
containing following attributes:
Qualification, Experience, Way of Problem solving,
Knowledge level, Interaction with guide, Journals published,
Implementation of algorithm, Relating with real-life
applications, Assessment, Punctual and Nature
i. Qualification (GRADUATE,
POSTGRADUATE, DOCTORATE)
ii. Experience (LESS_THAN 2, 2-4, 4-8, 8-10, 10
ONWARDS)
iii. Way of Problem Solving (POOR, AVERAGE,
GOOD, EXCELLENT)
iv. Knowledge level (POOR, AVERAGE, GOOD
EXCELLENT)
v. Interaction with guide (POOR, AVERAGE
GOOD, EXCELLENT)
vi. Journals published (LESS_THAN 1, 2-4, 5
ONWARDS)
vii. Implementation of algorithm (YES, NO)
viii. Relating with real-life applications (YES, NO)
ix. Assessment (YES, NO)
x. Subject Knowledge (POOR, AVERAGE, GOOD,
EXCELLENT)
xi. Punctual (RARE, SOMETIMES, ALWAYS)
xii. Nature (COURTEOUS, RUDE,
INDIFFERENT)
The outcome class is: In-general-view (POOR,
GOOD, EXCELLENT). Here, we have converted the
continuous attributes to the discrete/categorical attributes by
considering the particular range as a class for simplicity and
applicability of ID3 algorithm.
Figure 1: Decision Tree
4. DETERMINING GUIDES VIEW AND
PROVIDING GUIDELINES TO SCHOLARS:
We solve the above mentioned problem using ID3 Algorithm.
To solve this, a decision tree is formed by classifying the
training data and then the outcome class value is determined.
The steps involved can be described as follows:
Decision Tree Construction: For each scholar registered in
university, we can have collective feedback for the attributes
enlisted in the problem statement. By using, ID3 algorithm, a
decision tree is formed by classifying the training data and
then the outcome class value is determined. The outcome class
will be the leaf node of the tree and the attribute values will be
the internal nodes and the arcs connecting the nodes are the
decision trees made during the decision tree construction.
Determination of in-general-view about scholar: If the
attribute values are provided, the decision tree formed after
classification can be used to determine the outcome class, by
traversing the tree using the attribute value. Scholars can
provide the attribute values to the constructed tree and obtain
outcome class value for self-evaluation.
Guidelines to the improvement of Scholars: Production
rules can be directly obtained by traversing from root to the
leves of the tree is the advantage of using decision tree. By
using the production rules, we can provide the guidelines for
the improvement of scholar.
For example: If a scholar gets ‘Poor’ as outcome class value
due to less value for some attribute(s), we can also give the
ways to get the outcome class value as ‘Excellent’, such as
values for Regularity attribute should be ALWAYS instead of
SOMETIMES, etc.
Thus the scholar can improve according to the guidelines. The
outcome will be more accurate, when the training set is larger.
If the training data set is too small, then it may not consider all
the possibilities for the particular outcome and the result may
not be accurate.
CONCLUSIONS AND FUTURE WORK
We conclude that ID3 Algorithm works well on classification
problems. In this paper, we use decision tree algorithm to
classify the dataset obtained from Guides feedback. We
determine guides in-general-view about scholars and also
provide guidelines to the scholars. This will be helpful for
scholars to evaluate themselves and to improve accordingly.
This will find its applicability in scholars’ assessment process.
In future, we are trying to implement with software tools and
we will assess the attribute values and calculate outcome class
by getting input values from professors in universities.

__________________________________________________________________________________________
REFERENCES
[1] Sonika Tiwari and Prof. Roopali Soni,’ Horizontal
partitioning ID3 algorithm A new approach of detecting
network anomalies using decision tree’, International Journal
of Engineering Research & Technology (IJERT)Vol. 1 Issue
7, September – 2012.
[2] Anand Bahety,’ Extension and Evaluation of ID3 –
Decision Tree Algorithm’ University of Maryland, College
Park
[3] M. H. Dunham, DATA MINING: Introductory and
Advanced Topics, Pearson Education, Sixth Impression, 2009.
[4] P. Ozer, Data Mining Algorithms for Classification, BSc
Thesis, Artificial Intelligence, January 2008.
[5] A. Bahety, Extension and Evaluation of ID3 – Decision
Tree Algorithm
[6] J. R. Quinlan, Simplifying decision trees, International
Journal of Man-Machine Studies, 27, 221-234, 1987.
[7] P. H. Winston, Artificial Intelligence, Third Edition
Addison- Wesley, 1992.

Research scholars evaluation based on guides view

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Research scholars evaluation based on guides view (20)

More from eSAT Publishing House (20)

Recently uploaded (20)

Research scholars evaluation based on guides view