05-12-2012, 06:36 PM
Data Mining Algorithms to Classify Students
1Data Mining Algorithms.pdf (Size: 174.36 KB / Downloads: 22)
Abstract.
In this paper we compare different data mining methods and
techniques for classifying students based on their Moodle usage data and the
final marks obtained in their respective courses. We have developed a specific
mining tool for making the configuration and execution of data mining
techniques easier for instructors. We have used real data from seven Moodle
courses with Cordoba University students. We have also applied discretization
and rebalance preprocessing techniques on the original numerical data in order
to verify if better classifier models are obtained. Finally, we claim that a
classifier model appropriate for educational use has to be both accurate and
comprehensible for instructors in order to be of use for decision making.
Introduction
The ability to predict/classify a student’s performance is very important in web-based
educational environments. A very promising arena to attain this objective is the use of
Data Mining (DM) [26]. In fact, one of the most useful DM tasks in e-learning is
classification. There are different educational objectives for using classification, such as:
to discover potential student groups with similar characteristics and reactions to a
particular pedagogical strategy [6], to detect students’ misuse or game-playing [2], to
group students who are hint-driven or failure-driven and find common misconceptions
that students possess [34], to identify learners with low motivation and find remedial
actions to lower drop-out rates [9], to predict/classify students when using intelligent
tutoring systems [16], etc. And there are different types of classification methods and
artificial intelligent algorithms that have been applied to predict student outcome, marks
or scores. Some examples are: predicting students’ grades (to classify in five classes: A,
B, C, D and E or F) from test scores using neural networks [14]; predicting student
academic success (classes that are successful or not) using discriminant function analysis
[19]; classifying students using genetic algorithms to predict their final grade [21];
predicting a student’s academic success (to classify as low, medium and high risk classes)
using different data mining methods [30]; predicting a student’s marks (pass and fail
classes) using regression techniques in Hellenic Open University data [18] or using
neural network models from Moodle logs [11].
Background
Classification is one of the most frequently studied problems by DM and machine
learning (ML) researchers. It consists of predicting the value of a (categorical) attribute
(the class) based on the values of other attributes (the predicting attributes). There are
different classification methods, such as:
- Statistical classification is a procedure in which individual items are placed into groups
based on the quantitative information of characteristics inherent in the items (referred
to as variables, characters, etc.) and based on a training set of previously labelled items
[23]. Some examples of statistical algorithms are linear discriminant analysis [21],
least mean square quadratic [27], kernel [21] and k nearest neighbors [21].
- A decision tree is a set of conditions organized in a hierarchical structure [25]. It is a
predictive model in which an instance is classified by following the path of satisfied
conditions from the root of the tree until reaching a leaf, which will correspond to a
class label. A decision tree can easily be converted to a set of classification rules. Some
of the most well-known decision tree algorithms are C4.5 [25] and CART [4].
- Rule Induction is an area of machine learning in which IF-THEN production rules are
extracted from a set of observations [11]. The algorithms included in this paradigm can
be considered as a heuristic state-space search. In rule induction, a state corresponds to
a candidate rule and operators correspond to generalization and specialization
operations that transform one candidate rule into another. Examples of rule induction
algorithms are CN2 [8], AprioriC [17], XCS [32], Supervised Inductive Algorithm
(SIA) [31], a genetic algorithm using real-valued genes (Corcoran) [10] and a
Grammar-based genetic programming algorithm (GGP) [15].
Moodle Data Mining Tool
We have developed a specific Moodle data mining tool oriented for use by on-line
instructors. It has a simple interface (see Figure 1) to facilitate the execution of data
mining techniques. We have integrated this tool into the Moodle environment itself. In
this way, instructors can both create/maintain courses and carry out all data mining
processing with the same interface. Likewise, they can directly apply feedback and
results obtained by data mining back into Moodle courses. We have implemented this
tool in Java using the KEEL framework [1] which is an open source framework for
building data mining models including classification (all the previously described
algorithms in Section 2), regression, clustering, pattern mining, and so on.
Experimental Results
We have carried out some experiments in order to evaluate the performance and
usefulness of different classification algorithms for predicting students’ final marks based
on information in the students’ usage data in an e-learning system. Our objective is to
classify students with equal final marks into different groups depending on the activities
carried out in a web-based course. We have chosen the data of 438 Cordoba University
students in 7 Moodle courses (security and hygienee in the work, projects, engineering
firm, programming for enginnering, computer science basis, applied computer science,
and scientific programming). Moodle (http://moodle.org) is one of the most frequently
used free Learning Content Management Systems (LCMS). Moodle keeps detailed logs
of all activities that students perform in a data base. Information is available about the use
of Moodle activities and resources (assignments, forums and quizzes). We have
preprocessed the data in order to transform them into a suitable format to be used by our
Moodle data mining tool. First, we have created a new summary table (see Table 1)
which integrates the most important information for our objective (Moodle activities and
the final marks obtained in the course). Using our Moodle mining tool a particular
teacher could select these or other attributes for different courses during the data
preprocessing phase. The Table 1 summarises row by row all the activities done by each
student in the course (input variables) and the final mark obtained in this course (class).
Conclusions
In this paper we have compared the performance and usefulness of different data mining
techniques for classifying students using a Moodle mining tool. We have shown that
some algorithms improve their classification performance when we apply such
preprocessing tasks as discretization and rebalancing data, but others do not. We have
also indicated that a good classifier model has to be both accurate and comprehensible for
instructors. In future experiments, we want to measure the compressibility of each
classification model and use data with more information about the students (i.e. profile
and curriculum) and of higher quality (complete data about students that have done all the
course activities). In this way we could measure how the quantity and quality of the data
can affect the performance of the algorithms. Finally, we want also test the use of the tool
by teachers in real pedagogical situations in order to prove on its acceptability.