Intelligent Student Profiling System

Surname 1

Student Name

Professor

CS345

Date

Recommendation System: Intelligent Student Profiling System

ABSTRACT

This paper presents a proposal for an intelligent student profiling system using the knowledge of

a recommendation system. A recommendation system helps to give a user a number of filter

algorithms to provide a user with suggestions tailored to an individual. They help track the user’s

actions and suggest what they may like or not like in a list of items. Such recommendation systems

seek to provide predictions based on filtering a huge amount of data. This kind of filtering involves

data mining algorithms. Intelligent student filing system is a system that can be incorporated in a

learning process to enable effective learning. Most traditional web-based systems have wide ranges

of drawbacks when they are being compared with the real life classroom teaching situation. Most

of them lack a flexible support which is not collaborative with students. An effective student

profiling system will enable personalization of the student data and will enhance learning

motivation of students. Therefore, through the use of recommendation systems technique of

filtering noise data of student we can achieve the development of an intelligent system which can

help predict the student performance through data personalization. The concept employed here is

educational data mining which allows the finding of new patterns in large amounts of data. Hence,

extraction of wide knowledge from raw data offering interesting possibilities for education domain

(Carlsen et al.)

Surname 2

PROBLEM STATEMENT

Due to lack of an adaptive, flexible and collaborative support lacking in the many familiar web-

based education systems, the intelligent student profiling systems seeks to bridge this gap through

personalization of student data. The system will allow students to get personalised advice, extra

exercise questions and even personalised learning materials. Further, the students will be able to

obtain appropriate guidance and several functionalities from the system to motivate their learning

process.

SCOPE

This system will use a dataset created from a database which composed of school reports. The

reports are based on a paper sheet and other attributes like the number of school absences and four-

period grade. The dataset will be integrated into two subjects so that to enable a faster sampling

and filtering of the data before populating the system with other subjects.

JUSTIFICATION

The fastest growing areas in education currently is the use of web-based education systems. It has

formed basis for various research based in education technology. These systems are independent

of teaching and learning when compared to time and space flexibility. With the many systems

existing in the market, most of them do not understand the student model and hence they do not

respond appropriately to student activities. Through data mining and filtering of this noise data

from students, the intelligent student profiling system will be of value in bridging this gap in

comparison to real world classroom situation. An appropriate incorporation of technology will

result in the effective learning process with improved student performance. The system will

provide an enhanced support to instructors in arranging a real-life classroom set up.

Surname 3

OBJECTIVES

The main objective of developing intelligent student profiling system is to ensure a collaborative

education system which provides students with personalised learning based on the student

capabilities and interests. Also, it is to develop a system that understands the real life classroom

situation and provide personalised extra question exercises and personalised advice. Further, it

provides a system with incentive functionalities to motivate the student learning process. Also, to

provide instructors with a system that will reduce the burden of grouping students in teams and

they can understand the appropriate advice to offer the students.

BACKGROUND OF THE STUDY

This work is a motivation as a result of research related to education data mining and involvement

of technology in the learning process. It is a one way to illustrate how ensemble techniques and

filtering algorithms can be applied in the context of supervised learning with the aim of enhancing

accuracy and stability in the student performance prediction process. This work herein is a

presentation of application of a hybrid of these techniques so that to enable instructors to predict

the students’ performance and then they can be able to group them in teams for further advising

and personalised learning (Dutt, Ashish). Through the use of data mining and analysis algorithms,

we can achieve the goal of developing this intelligent student profiling system. These algorithms

provide the basic required in data filtering and clustering.

LITERATURE REVIEW

Data mining is the process of exploration and analysis of large quantities of data with the aim of

discovering meaningful patterns and rules. This is an interdisciplinary field in computer science

which involves a computational process of large data sets' pattern discovery. This extraction of

Surname 4

knowledge from a large data set of information and transform it into an understandable structure

which can be used further in making decisions and in making predictions is the main aim of data

mining. It involves solving problems by analysing data already present in databases. Also, it is an

essential process of application of intelligent methods to extract important data patterns from noise

data with the aim of improving the outcome (Luan, Jing)

Recommendation systems have been used in various fields to suggest a list of items from the

existing items in databases so that to increase more like or dislikes of a user. To develop a

recommendation system it invokes the use of very collaborative filtering and ratings. For instance,

ratings can be obtained through monitoring of the users' actions directly. This kind of filtering have

been employed in these various fields and have showcased positive outcomes. For example, they

are used in the development of scholarly search engines like Google Scholar, where the focus is

on classic text mining and citations counts on the web. All these recommendation systems use

filtering algorithms which are important in analysis. The systems can be of different categories

depending on the objectives. They include web recommendation applications, information

recommendation domain, information filtering and sharing domain and much more (Adomavicius,

G., and A. Tuzhilin)

LIMITATIONS

The main challenge of intelligent student profiling system is that it is very difficult to suggest

recommendations to new users. This is because the profile is empty and hence, the taste or

preferences of the user are unknown to the system. This is referred to as cold start problem. To

solve such problem one needs to employ hybrid approaches. Also, another challenge which most

recommendation system face is the issue of privacy. The system collects much of the user’s

Surname 5

information. Naturally, most users wonder with their privacy, security, and confidentiality of their

information in the system.

TIME SCHEDULE

Duration

Activity

Week 1-2

Identification of the project title

Week 3-4

Preparation of project requirements

Week 5-8

Design of the system interface

Week 9-10

Coding of the project

Week 11-12

Testing of system modules and integration

Week 13

Testing the whole system and implementation

ALGORITHM

Most recommendation systems are developed based on machine learning techniques. This involves

the application of data mining and the making of statistical inferences from the existing data. This

approach requires that the system is provided with data from the past from which the algorithm

will learn to predict the future outcomes. The system will employ K-means clustering algorithm

and ensemble algorithm. It might even invoke the use of model-based approaches like regression

approach. Also, at some point, students are supposed to rate some subjects so that it can be easy to

determine those who have same interests for effective clustering. However, there are several

clustering algorithms which can be applied to educational data sets in understanding the emotional

intelligence of students. Through data mining, we can identify, extract, and evaluate variables in

relation to the learning process of students. Through the collaboration of K-means clustering

Surname 6

algorithm and Bayes classification and ensemble filtering, we can filter and have a clear prediction

of student performance (Sarma, T. Hitendra et al.)

CONCLUSION

It is clear that student data can be filtered to show a huge improvement in predictive accuracy.

Student information can be clustered into well-defined groups in relation to their learning patterns

after filtering of the noisy raw data and the outcome results are good. Therefore, the system will

allow instructors to have an easy time in advising, tutoring and forming class teams because the

exercise is made easier through grouping.

RECOMMENDATION

This work also presents some areas which need further research on them so that learning process

can be easy and effective to both students and instructors. The areas which need further research

include the one listed in system limitations. For example a study on how to handle cold start for

new users in the system.

Surname 7

Work Cited

Adomavicius, G., and A. Tuzhilin. "Using Data Mining Methods To Build Customer Profiles".

Computer, vol 34, no. 3, 2001, pp. 74-82. Institute of Electrical And Electronics Engineers

(IEEE), doi:10.1109/2.901170.

Carlsen, Roger, and Dee Anna Willis. Society For Information Technology & Teacher Education

International Conference Annual. Chesapeake, Va, Association For The Advancement Of

Computing In Education, 2007.

Dutt, Ashish. "Clustering Algorithms Applied In Educational Data Mining". International

Journal of Information and Electronics Engineering, 2015, Ejournal Publishing,

doi:10.7763/ijiee.2015.v5.513.

Luan, Jing. "Data Mining And Its Applications In Higher Education". New Directions For

Institutional Research, vol 2002, no. 113, 2002, pp. 17-36. Wiley-Blackwell, doi:10.1002/ir.35.

Sarma, T. Hitendra et al. "Speeding-Up The Kernel K-Means Clustering Method: A Prototype

Based Hybrid Approach". Pattern Recognition Letters, vol 34, no. 5, 2013, pp. 564-573.

Elsevier BV, doi:10.1016/j.patrec.2012.11.009.

Place new order. It's free, fast and safe

Paper type

Deadline

Page count

550 words

Total price: $ 15.00

Our customers say

Jeff Curtis

USA, Student

"I'm fully satisfied with the essay I've just received. When I read it, I felt like it was exactly what I wanted to say, but couldn’t find the necessary words. Thank you!"

Ian McGregor

UK, Student

"I don’t know what I would do without your assistance! With your help, I met my deadline just in time and the work was very professional. I will be back in several days with another assignment!"

Shannon Williams

Canada, Student

"It was the perfect experience! I enjoyed working with my writer, he delivered my work on time and followed all the guidelines about the referencing and contents."

Intelligent student profiling system

Place new order. It's free, fast and safe

Our customers say

Jeff Curtis

Ian McGregor

Shannon Williams