DEVELOPMENT OF SOFTWARE SYSTEM FOR DATA MINING OF STUDENTS' SURVEYS FOR ASSESSMENT OF COURSES TEACHING QUALITY IN ENGINEERING COLLEGE OF BAGHDAD
|
INTRODUCTION 4
L REVIEW OF RELATED WORK 8
1.1. Overview of the data mining techniques 8
1.2. Papers on educational data mining 11
1.3. Modem sortware to perform data mining 17
2. DESIGN OF EDUCATIONAL DATA MINING SYSTEM . . 19
2.1. Basic principles...... t 19
2.2. Designing of structure and behavior of
the educational mining system 19
2.3. Design of quizzes to gather
educational data 21
2.4. Design of web-site that will implement quizzes
for evaluation process . * •... * . .. ■ , ♦ .,. ...., , . ♦.+ 23
2.5. Design of scenarios of applying data mining
technique 27
3. IMPLEMENTATION 31
4. EXPERIMENTS AND DISCUSSION 38
CONCLUSIONS 45
REFERENCES 46
L REVIEW OF RELATED WORK 8
1.1. Overview of the data mining techniques 8
1.2. Papers on educational data mining 11
1.3. Modem sortware to perform data mining 17
2. DESIGN OF EDUCATIONAL DATA MINING SYSTEM . . 19
2.1. Basic principles...... t 19
2.2. Designing of structure and behavior of
the educational mining system 19
2.3. Design of quizzes to gather
educational data 21
2.4. Design of web-site that will implement quizzes
for evaluation process . * •... * . .. ■ , ♦ .,. ...., , . ♦.+ 23
2.5. Design of scenarios of applying data mining
technique 27
3. IMPLEMENTATION 31
4. EXPERIMENTS AND DISCUSSION 38
CONCLUSIONS 45
REFERENCES 46
Topicality of The Research
Data mining is the process of discovering interesting knowlege, such as patterns, association, changes, anomalies and significant structures, from large amounts of data stored in database, data warehouses, or other information repositories (11]. In recent years, there has been increasing interest in the use of data mining to investigate scientific questions within educational research, an area of inquiry termed educational data mining. Educational data mining (also referred to as “EDM”) is defined as the area of scientific inquiry centered around the development of methods for making discoveries within the unique kinds of data that come from educational settings, and using those methods to better understand students and the settings which they leam in [5].
E-leaming system provides data (hat is transformed into information used by teachers in order to improve leaching process. Students also benefit from that in formal ion, and also by adapting their learning process. The goal is to understand the learning and teaching in order to improve its quality.
El )M might be irvil Ini’ ^indent modeling and e-learning system wfiliation, Teachers group or classify students, discover patterns of misconceptions, etc.
In classroom, teachers observe students behavior and analyze lest results. They adapt the instruction according to the provided feedback. Students are different and respond to the changes according to their characteristics, so teacher constantly adapts. Such information and feedback is often missing in e-learning systems.
Since such systems store large amount of data, EDM techniques might provide meaning to the data as basis for adaptation, scheduling, prediction of students ’dropout, or course enrollment.
In general, EDM process consists of three phases: data collection and preparation, data analysis and interpretation of results [18).
Background
Student end-of-course evaluations are widely used by colleges and universities io determine the factors that help success of courses and the effectiveness of their instructors, also to define the factors or attributes arc often appear together in good and bad courses.
The success of the assessment depends on how this questionnaire has real, accurate and focused details questions about each semester and the teacher associated with it. this helps the head of department to restructure coursework by giving importance to some of the attributes have obvious impact on the final assessment and was ignored previously.
As well as in terms of final evaluation of the teaching, it gives an opportunity for each instructor to identify his weaknesses front the student’s point of view, while making the head of the department or college to reconsider the appropriate instructor depending on the specialization and experience.
In our research, we look the College of Engineering at the University of Baghdad, which consists of 12 scientific departments, where the focus was on the Department of Civil Engineering, which is the first department opened in the university and includes many engineering branches: construction engineering, engineering bridges and roads, environmental engineering, cie.
The duration of study in this college is four years where each year aboui 7 semesters have studied and these semesters are divided according to their importance to basic and secondary classes. Therefore, the number of lectures per year is differenced and some of these semesters contain a theoretical and practical part,
We took an assessment of approximately 19 semesters at a rale of 400 students divided by the four grades, and the models were divided between the main and secondary classes, containing a practical lesson or not.
For these students, we noticed a significant decrease in the final grades and the attendance in the lectures is relatively low. Therefore, college managc- ment needs to Proper understanding on the learning behavior of the students helps the educational programs, to a much more improved level, which can increase the learning capabilities of the student, who followed their educational programs.
1 lore came (he benefit of the use various data mining methodologies in different ways to identify learning pattern of student.
Research goal and objectives
The goal of the research is the evaluation of the semester courses and their instructors in the Engineering College, This evaluating is based on many factors and attributes that used as questions, which arc answered by the students of that college.
For the reaching this goal we must solve the following objectives.
1. Overview the data mining techniques and papers on educational data mining.
2. Design quizzes to ask students about the courses and instructors, and develop an application for performing quizzes and collecting educational data
3. Design educational data mining scenarios for understanding of the causes for the problems above.
4. Implement the developed scenarios as Data Mining system for discovering hidden knowledge.
5. Implement the developed scenarios as Data Mining system for discovering hidden knowledge.
6. Apply the developed Data Mining system to the collected data and present the extracted knowledge lo the College authorities.
Outline of The Research
This thesis consists of the introduction, four chapters, conclusions and a list of references.
The first chapter contains oven icw of the data mining techniques classification, association Rules and clustering, a literature review on previous works in educational data mining and review of modem software to perform data mining.
Chapter 2 contains basic principles about design quizzes to gather data and the software that implemented them, also describe the scenarios that applied on processing collected data.
Chapter 3 contains implementation of software system.
Chapter 4 contains experiments and discussion of the developed software system.
Conclusion contains summarizing remarks on results of the thesis.The volume of the thesis is 4 У pages .The volume of the list of references is 33 sources.
Data mining is the process of discovering interesting knowlege, such as patterns, association, changes, anomalies and significant structures, from large amounts of data stored in database, data warehouses, or other information repositories (11]. In recent years, there has been increasing interest in the use of data mining to investigate scientific questions within educational research, an area of inquiry termed educational data mining. Educational data mining (also referred to as “EDM”) is defined as the area of scientific inquiry centered around the development of methods for making discoveries within the unique kinds of data that come from educational settings, and using those methods to better understand students and the settings which they leam in [5].
E-leaming system provides data (hat is transformed into information used by teachers in order to improve leaching process. Students also benefit from that in formal ion, and also by adapting their learning process. The goal is to understand the learning and teaching in order to improve its quality.
El )M might be irvil Ini’ ^indent modeling and e-learning system wfiliation, Teachers group or classify students, discover patterns of misconceptions, etc.
In classroom, teachers observe students behavior and analyze lest results. They adapt the instruction according to the provided feedback. Students are different and respond to the changes according to their characteristics, so teacher constantly adapts. Such information and feedback is often missing in e-learning systems.
Since such systems store large amount of data, EDM techniques might provide meaning to the data as basis for adaptation, scheduling, prediction of students ’dropout, or course enrollment.
In general, EDM process consists of three phases: data collection and preparation, data analysis and interpretation of results [18).
Background
Student end-of-course evaluations are widely used by colleges and universities io determine the factors that help success of courses and the effectiveness of their instructors, also to define the factors or attributes arc often appear together in good and bad courses.
The success of the assessment depends on how this questionnaire has real, accurate and focused details questions about each semester and the teacher associated with it. this helps the head of department to restructure coursework by giving importance to some of the attributes have obvious impact on the final assessment and was ignored previously.
As well as in terms of final evaluation of the teaching, it gives an opportunity for each instructor to identify his weaknesses front the student’s point of view, while making the head of the department or college to reconsider the appropriate instructor depending on the specialization and experience.
In our research, we look the College of Engineering at the University of Baghdad, which consists of 12 scientific departments, where the focus was on the Department of Civil Engineering, which is the first department opened in the university and includes many engineering branches: construction engineering, engineering bridges and roads, environmental engineering, cie.
The duration of study in this college is four years where each year aboui 7 semesters have studied and these semesters are divided according to their importance to basic and secondary classes. Therefore, the number of lectures per year is differenced and some of these semesters contain a theoretical and practical part,
We took an assessment of approximately 19 semesters at a rale of 400 students divided by the four grades, and the models were divided between the main and secondary classes, containing a practical lesson or not.
For these students, we noticed a significant decrease in the final grades and the attendance in the lectures is relatively low. Therefore, college managc- ment needs to Proper understanding on the learning behavior of the students helps the educational programs, to a much more improved level, which can increase the learning capabilities of the student, who followed their educational programs.
1 lore came (he benefit of the use various data mining methodologies in different ways to identify learning pattern of student.
Research goal and objectives
The goal of the research is the evaluation of the semester courses and their instructors in the Engineering College, This evaluating is based on many factors and attributes that used as questions, which arc answered by the students of that college.
For the reaching this goal we must solve the following objectives.
1. Overview the data mining techniques and papers on educational data mining.
2. Design quizzes to ask students about the courses and instructors, and develop an application for performing quizzes and collecting educational data
3. Design educational data mining scenarios for understanding of the causes for the problems above.
4. Implement the developed scenarios as Data Mining system for discovering hidden knowledge.
5. Implement the developed scenarios as Data Mining system for discovering hidden knowledge.
6. Apply the developed Data Mining system to the collected data and present the extracted knowledge lo the College authorities.
Outline of The Research
This thesis consists of the introduction, four chapters, conclusions and a list of references.
The first chapter contains oven icw of the data mining techniques classification, association Rules and clustering, a literature review on previous works in educational data mining and review of modem software to perform data mining.
Chapter 2 contains basic principles about design quizzes to gather data and the software that implemented them, also describe the scenarios that applied on processing collected data.
Chapter 3 contains implementation of software system.
Chapter 4 contains experiments and discussion of the developed software system.
Conclusion contains summarizing remarks on results of the thesis.The volume of the thesis is 4 У pages .The volume of the list of references is 33 sources.
In this research, we presented our methodology and results of analyzing a large set of undergraduate course evaluations from an Engineering faculty. We have shown how useful the application of data mining techniques in course management systems, although wc have shown these techniques separately, they can also be applied together in order to obtain interesting information in a more efficient and faster way.
Based on our analysis, in order to improve the teaching quality, instructors should consider enhance their attitude, and organization visual presentation skills. They want to make sure that they respond questions well and clearly. In order to improve the course quality, instructors may want to design tests and assignments such that they are closely related to the course material.
Due to the low evaluations on the usefulness of texthooks and tutorials, instructors may consider improving the quality of texthook and tutorials. From an institution's perspective, it should try to schedule more morning classes with smaller number of students since evening classes and classes with large size receive worse evaluations.
Also, we applied clustering techniques to obtain the exact groups of students can be divided into. And these groups can also he used to create a classifier in order to classify students. The classifier shows what the main characteristics of the students in each group are, and it allows new online students to be classified. Finally, we applied association rule mining on the most important deterministic of the course and found when the type of the course is essential this mean more theoretical part and less practical aspect with many workload on students,leads to less final success ratio and more absence from the lectures.
Based on our analysis, in order to improve the teaching quality, instructors should consider enhance their attitude, and organization visual presentation skills. They want to make sure that they respond questions well and clearly. In order to improve the course quality, instructors may want to design tests and assignments such that they are closely related to the course material.
Due to the low evaluations on the usefulness of texthooks and tutorials, instructors may consider improving the quality of texthook and tutorials. From an institution's perspective, it should try to schedule more morning classes with smaller number of students since evening classes and classes with large size receive worse evaluations.
Also, we applied clustering techniques to obtain the exact groups of students can be divided into. And these groups can also he used to create a classifier in order to classify students. The classifier shows what the main characteristics of the students in each group are, and it allows new online students to be classified. Finally, we applied association rule mining on the most important deterministic of the course and found when the type of the course is essential this mean more theoretical part and less practical aspect with many workload on students,leads to less final success ratio and more absence from the lectures.
