Intelligent Tutoring System for Learning Object-Oriented Programming Language
Chapter One
Aim and Objectives
Aim
This project aims to develop a solution, which is an Intelligent Tutoring System (ITS) for learning an Object Oriented Programming language (OOP), case study, Java programming language, so as to help in enhancing students’ learning and provide adaptive learning to each student.
Objectives
The main objectives of this research work are:
- Give a practical and vivid explanation of the main components or architecture of an ITS system and how it
- To propose and develop an Intelligent Tutoring System (ITS) for learning an Object Oriented Programming Language (OOP), case study, Java programming language, which will hopefully provide an efficient solution or a better approach towards providing personalized and adaptive learning to students without requiring any human intervention.
CHAPTER TWO
LITERATURE REVIEW
This chapter presents literature reviews that have been done in regards to major aspects where Intelligent Tutoring System (ITS) has been applied.
Introduction
Adaptive learning and personalised learning represents a key topic with regards to Intelligent Tutoring Systems (ITSs). The ability of ITS to adapt their environment to the needs of the students is often recalled as one of the explanations to support the implementation of ITS within learning projects.
It seems vital to emphasize on student’s personalized learning. This is the reason why an ITS requires a good profiling procedure, an activity that leads to the composition of a user profile able to collect and elaborate information that are considered important for the recognition of specific needs.
User profiles are built referring to different models that focus on various characteristics of the individuals. The data considered can vary according to the particular research hypothesis, even according to the specific learning outcomes that are associated with the ITS.
The aim of this paper is to outline a detailed overview on the main progresses made in the field of user modelling and user profiling. We will take into account both the traditional approaches and the recent advantages, in order to highlight new lines of research.
Different types of ITS from different types of personalization procedures
Different types of ITS can be classified thanks to various characteristics. An alternative way to diversify them is to consider the various methodologies that are the basis of the personalization process.
Cognitive Tutors (CT)
The first family of ITS is that of Cognitive Tutors (CT). They derive from an approach that is grounded in a specific theoretical basis: the ACT-R theory of cognition (Adaptive Control of Thought-Rational), developed by John Anderson (Anderson, 1990; Anderson et al., 1994).
The ACT-R theory formulates a representation of human knowledge, dividing it into declarative knowledge and procedural knowledge. Declarative knowledge is a factual knowledge, involving facts, images or sounds. It is typically acquired through perception and its elements are usually considered as chunks of knowledge (Mitrovic et al., 2003). Procedural knowledge is goal-oriented, because it is related to the comprehension of how to do things. It is typically acquired through practice and its elements are usually conceptualized as production rules: they specify the plausible conditions of their application and the consequences, in terms of actions, of their application (Ritter et al., 2007).
The learning process arises from two phases. It starts accumulating declarative knowledge, while in a second phase it is converted in declarative knowledge. A complex task involves a set of cognitive skills to be resolved. Such skills can be decomposed and represented as production rules. The ability to solve a given task, therefore, lies in the control of those elements of knowledge.
CT promises personalized support while students are engaged in problem solving activities. The system observes each learner’s behaviour to identify the most suitable problems to assign and the most suitable feedback to provide. They can achieve this goal using a cognitive model that represents the description of the skills and strategies, expressed as a set of production rules, required to solve a task in a particular knowledge domain. Creating such a model is a complex challenge, because it must include all the knowledge components that are considered essential within a certain domain, the analysis of human behaviour while solving a particular problem, all the possible paths that we can follow to solve it. (Ritter et al, 2007).
A CT constantly monitors its users, collecting information about their behaviour in a profile. Then, it compares it with the elements stored in the cognitive model in order to assess if a student needs help. This process is called model tracing. If an inappropriate action is detected the tutor reports it to the student and gives him/her suitable hints and feedback. Another important process is that of knowledge tracing: each action of the student is related to one of the skills provided in the cognitive model (Aleven & Koedinger, 2002). Then, the tutor makes a prediction about the probability that this skill is correctly mastered by the learner. The tutor uses this information to propose activities and exercises that focus the attention on those skills that must be strengthened (Ritter at al., 2007).
Constraint Based Models (CBM)
The second family of ITS is usually defined Constraints Based Models (CBM). They are projected to overcome some limitations of CT.
To design cognitive models, we need a long and very complex process. Referring to some particular domains, such as Human Sciences, the process can result impossible because problems can have multiple solutions and, furthermore, they can be solved following a wide variety of possible paths. Even referring to those domains, such as Mathematical Sciences, in which the process is possible, CT present a second type of problems. The main purpose of the tutor is to understand when and why a certain student makes an error. The tutor can understand it only if it is able to replicate that action, even if it is incorrect. As a consequence, it becomes essential to incorporate within its cognitive model also the so-called buggy rules, that are all those choices that lead to a wrong result (Mitrovic, 2012).
CHAPTER THREE
METHODS AND DESIGN
INTRODUCTION
In this chapter, we describe the methods used in carrying out the study and accumulating data for analysis and design. Before the development of an efficient Intelligent Tutoring System, hereinafter termed ITS, it is important to undertake a careful and analytical study also known as System analysis. The reason is to identify the inconsistency prone areas of ITS operations as well as the flaws affecting accuracy, reliability, data integrity, speed of operation, and overall efficiency of the system. This chapter encapsulates the following: Research design, Research intervention, Research instrument, Research Processes, Data Analysis and Interpretation. It also includes System architecture, Program design, File organization, System models, Programming Language, Interaction models, Database used, System control and System requirements.
Methodology
The method adopted here for program design is the structured software development approach using Object-Oriented Analysis, Design and Implementation. This is carefully done to ensure that the software can be read, understood and built upon in the future.
As it has been stated in Chapter 1, the problem that this research work tends to address is to enhance learning in a more efficient way. Traditionally, classrooms can be over- populated with students. In that kind of a condition, learning may not be efficiently administered thus, limiting understanding of most students.
Research Design
In this research work, we will make use of both the traditional waterfall approach to software development and incremental build model as the work is designed in a sequence from start to finish. (Ukem and Ofoegbu, 2012) referred to the Waterfall software life-cycle model as a traditional model because it was the first widely used software development life cycle. They stated that it is part of Structured Software Engineering or the Structured Paradigm, which is the older of the two earliest approaches to formal software engineering.
CHAPTER FOUR
IMPLEMENTATION AND RESULTS
This chapter discusses about the results and discussions. Screenshot is used in the presentation of our findings. This chapter includes implementation which comprises of the programming language used, text input, preprocessing, matching, and evaluation of the performance.
Presentation of Results
The solution was implemented in the Java programming language (Netbeans IDE 8.2) and run on a Hewllet Packard (HP 630) laptop @2.00GHz with 6GB RAM on Windows 10. Following on the development methodology, this chapter describes design implementation in terms of Program coding, Program user interface, Database design and code logic for the entire system.
CHAPTER FIVE
SUMMARY, CONCLUSION, RECOMMENDATIONS, AND FUTURE WORK
This chapter discusses the complete thesis report, which highlights the importance of our findings in summary, conclusion, recommendations, and the future work.
Summary and Conclusion
A computer system that aims to provide immediate and customized instruction or feedback to learners, usually without requiring intervention from a human teacher is known as an Intelligent Tutoring System.
ITS have four (4) main components as stated below:
- Communication module
- Domain module
- Learner’s module
- Pedagogical module
It is imperative to mention that an adaptive ITS needs to select the best instructional strategies at the right time, depending on the learner’s profile in certain conditions and the learning process in general. Selections of teaching strategies should be done in order to maximize in-depth learning and motivation so as to reduce costs and training time.
D1 is a diagram, which analyse students who have high access to an ITS (Student performance page). According to Markov, learner’s in this category tend to experience higher enjoyment and interest to learn. Thus, there is a zero probability of getting bored when a student is interested in a topic.
D2 on the other hand, shows students who have less access to an ITS (Student performance page). According to Markov, there is a level of probability that students in this group lose interest in a topic, thereby transiting into the bored-emotional state.
Recommendations
- Understanding learner’s cognitive ability is very important in any ITS Thus, more instructional mediums such as images, videos, Illustrations, etc. can be added to the ITS system.
- A chatting system can be added to the ITS system where learners can chat with otherlearners, share and discuss their learning tasks Hence, they will be an admin who will monitor the chats of the learners and the learners can further relay their queries to the admin.
- Providing adaptive learning to users is key in an ITS system. I suggest incorporating ways of determining user’s emotional status into the ITS
Future Work
As a future work, this research study can further be improved upon by the addition of facial and body emotional analysis to the ITS System.
It involves the following steps as can be seen in the diagram bellow:
REFERENCES
- Aleven, V. & Koedinger, K. R. (2000). Limitations of Student Control: Do Students Know when they need help? In Intelligent Tutoring Systems: 5th International Conference, ITS 2000, Montreal, Canada, June 19th-23rd, 2000, pp. 292-303.
- Aleven, V., McLaren, B., Roll, I., Koedinger, K. R. (2006). Toward Meta-cognitive Tutoring: A Model of Help-Seeking with a Cognitive Tutor. International Journal of Artificial Intelligence in Education 16 (2), pp. 101-130.
- Aleven, V. & Koedinger, K. R. (2002). An effective metacognitive strategy: learning by doing and explaining with a computer-based Cognitive Tutor. Cognitive Science 26, pp. 147-179.
- Anderson, J. R. (1990). The adaptive character of thought. Hillsdale, NJ: Erlbaum.
- Anderson, J. R., Bothell, D., Byrne, M. D., Lebière, C., Qin, Y. (2004). An integrated theory of the mind. Psycological review 111, pp. 1036-1060.
- Arroyo, I., Ferguson K., Johns J., Dragon T., Meheranian H., Fisher D., Barto, A., Mahadevan, S., Woolf, B. P. (2007). Repairing Disengagement With Non-Invasive Interventions. In Proceedings of the 13th International Conference on Artificial Intelligence in Education, AIED 2007, Los Angeles, Usa, July 9th-13th, 2007, pp. 195- 202.
- Arroyo, I., Cooper, D. G., Burleson, W., Woolf, B. P., Muldner, K., Christopherson, R. (2009). Emotion Sensors Go To School. In Proceedings of the 14th International Conference on Artificial Intelligence in Education, AIED 2009, Brighton, UK, July 6th- 10th, 2009, pp. 17-24.
