Intelligent Tutoring Systems - Case Study Example

Intelligent Tutoring System Report Intelligent Tutoring Systems (ITS) or Intelligent Computer- Aided Instruction (ICAI) is so useful in the fields of education, psychology and artificial intelligence. Within these fields, ITS is capable to transform education by creating a better learning environment to the learner or learner groups. The aim of ITS is to enable the learner to help the learner to deal appropriately with a wide range of knowledge. For example, Reva Freedman points out that the learning process or a project which is focusing on intelligence is able to generate solutions to complex problems. So, the aim of this process is equip the learner to cope up with these varied situations. “a project focusing on intelligence in the domain model may generate solutions to complex and novel problems so that students can always have new problems to practice on” When the learner is provided with multiple or novel ways to solve the same problem, the problem solving capacity and productivity will increase. (Freedman, 2000). ITS can benefit the learner in many ways. First of all, it helps the learner to learn faster and to translate the learning into improved performance. It is able to reduce the amount of time for achievement in a specific field. It helps the learner to choose the best learning experience from a variety of learning experiences. Rebecca S. Crowley, Elizabeth Legowski, Olga Medvedeva, Eugene Tseytlin, Ellen Roh, and Drazen Jukic points out that learning by doing is the best learning experience that a system can provide to a learner. “ITS support “learning by doing”—as students work on computer-based problems or simulations of real-world tasks, the system offers guidance and explanations, points out errors, and organizes the curriculum to suit the needs of that individual.” (Rebecca S et al, 2007). When the learner develops skills and correct mistakes by the help of the system and the real aim of an intelligent system is fulfilled. Another benefit is that the participant is capable to practice their skills in highly interactive learning situations and it deals with a wide range of knowledge. Moreover, it is better than traditional class-room training process. Reichgelt, H points out that the role of ITS in a class room must be of an assistant teacher. “For an ITS to be used in the classroom, it must take on the role of assistant to the teacher. Further what teachers predominantly need from an assistant is configurability at every knowledge level.” (Reichgelt H, 1990). Here, the teacher can seek the advice of the intelligent assistant and can rectify mistakes. Through an ITS, there is high chance to keep track of all students, error manipulation and to receive feedback. But it is evident that non intelligent but well designed systems can be excellent in educational field. To support the system for problem solving there are certain tools like spreadsheet, graphs and symbolic calculator. Non-intelligent programs or systems have a single way to tackle the problem, but intelligent programs possess flexible ways to carry out skills of the learner. To design an intelligent tutoring system is not so easy because it needs deep research to identify the problems that the learner face in a specific field of learning and the remedial measures to rectify those problems. The designer must identify the problems and must design suitable learning situations and must be designed in an attractive way. The system that is developed is aimed to result in a positive change in cognition of the learner. So it is essential to identify the mistake of the learner. Esma Aimeur, Claude Frasson and Hugo Dufort points out that an effective ITS must help the learner by pinpointing the most misunderstood concept. “An indicator has been developed that measures discord between the ideas, helping to pinpoint the concepts that are most likely to be misunderstood by the learner.” (Aimeur, Frasson & Dufort, 2000). The best way to make an intelligent tutoring system more usable is to connect it to a server from where learners can collect sufficient data. For example, to develop a games-based learning intelligent tutoring system for Physics, specific knowledge domain of Physics must be in mind. Goals: 1. Conceptual knowledge acquisition; 2. Development of problem solving skills. Learning objectives: 1. Understanding of Newton’s three laws; 2. Understanding of Gravitational acceleration; 3. Calculation of speed, acceleration and velocity; and 4. Measurement. The game is set in a 3D island, where the learner can control the movements of the main character, a racing car. The learner or the player can navigate the car from the starting point to the finishing point through the least number of roads possible. The tachometer displays force, acceleration and angle for the car to travel a specified distance. The learner can decide the road, force, acceleration, speed etc of the car. The importance of the game is that it is able to develop the meta-cognitive skills such as self help skills, decision making capacity and after completing the less difficult stage; the learner can choose more difficult stages for further scoring. Here, the learner can calculate the speed, acceleration and velocity of the car. Moreover, learner can calculate his own score and can observe his/ her own progress. Another knowledge based learning experience in ITS is Domain Model. The history of basic computer aided learning (CAL) is essential to know more about Domain model. Chris Mills and Barney Dalgarno points out that the knowledge about various domain models is helpful to know more about CAL systems. “To understand the various domain model implementations within intelligent tutoring we must acknowledge the history of basic computer aided learning (CAL) systems from which ITS have evolved.” (Mills & Dalgarno, 2007). The basic form of CAL contains simple type of multiple choice tests or quizzes. These questions were so simple that it was not easy to evaluate the domains of economics or math. For years these question, answer and feedback method was used for instruction and evaluation. The drawback of this system was that the domain knowledge of this system is limited to a set of hard-coded question/answer pairings. Moreover, the improvement level of the learner is limited because the learning experiences are limited to question/answer pattern. The Domain model is knowledge based and it is the mechanism used for querying this knowledge base. After defining the problem of some specific problem, Domain model finds out an automatic solution with the help of domain resource component. Another peculiarity of Domain model is that it uses hints to complete the exercise which can analyze the learner’s input and it can locate the step where the learner is situated. As earlier pointed out, it is limited to a set of hard-coded question/ answer pairings. Moreover there is no further scope for the cognitive development of the learner. It is a simple device with yes or no questions which are objective in nature and its output is predictable. The division of systems to intelligent and non-intelligent is that the former lets the learner to interact with the system and helps the learner by providing valuable feedback. Eric A. Domeshek; Bruce W. Knerr; and William R. Howse points out that it is essential for a learner to be proficient in his domain field. If it is professional level skills or commanding level skills, there is no change. “Achieving expert levels of proficiency in professional-level reasoning skills-whether for battlefield commanders or for professionals in a wide range of other fields-requires extensive practice, coaching, and feedback.” (Domeshek, Knerr & Howse, 2008). So, to get feedback is also essential to develop one’s own skills. One can see that non-intelligent systems cannot provide effective feedback to the user. Moreover, they are only user interfaces where the learner can know his knowledge level without further feedback and its interaction is limited with the learner. So it can be seen that the power of interaction is the most important feature of an intelligent system. Another knowledge based learning experience in ITS is Learner model. The motive behind the construction of Learner model is to inspect individual and group performance through different ways in learning. Further, the learner can decide or choose his/ her own way and success. It is evident that there are two types of learners. One is active learners and other is reflective learners. Active learners like to do something active with the information but reflective learners prefer to think twice before acting. In Learner model, there is enough space to accommodate both these learners. Different individuals tackle the same problem by different ways and this is because of individual difference. So, all the learners may not benefit equally from Learner model. The learner model is so useful in my intelligent tutoring system because the aim of my gaming system is to interact with more learners and to collect the data and evaluate their performance. Only the learner model helps to identify the problem of the students because it is not objective but subjective. Moreover, it allows the learner to control the interactive system and helps in further improvement. When system domain model is considered, it lacks subjectivity. So, there is less chance for a learner to develop own skill. Here, James Ong and Sowmya Ramachandran I of the opinion that ITS systems are not like other systems but it ensures the skill development of the learner through interaction. “Unlike other computer-based training technologies, ITS systems assess each learner's actions within these interactive environments and develop a model of their knowledge, skills, and expertise.” (Ong & Ramachandran, 2003, p.2). So the assessment of the actions of the learner helps the ITS system to learn more about the learner and to develop a model according to their needs. These are the adaptive teaching or guidance decisions that the Domain and Learner models can provide to the system of learning. Learning become faster and better, increase productivity of the learner, reduce the amount of time for achievement, variety of learning experiences for the learner, and it can help the learner to solve a large number of problems within a limited time. Byung-In Cho points out that the selected problems in an intelligent system is aimed to raise problem solving capacity of the learner. So the selected problem must be challenging. “The selected problem must be challenging, but not frustrating. The problems should be varied to maintain the student's interest and to ensure coverage of important material.” Moreover, the problem should be able to maintain the interest of the learner. (Cho, 2000). The games-based learning intelligent tutoring system for Physics is an attempt to individualize the educational experiences according to the level of knowledge and skill in Physics. To design an intelligent tutoring system for an easier subject is just wastrel. Moreover, high ability subjects are best suited for ITS. The potential educational benefits of games-based learning experience are motivation, evaluation of progress and acquisition of problem solving techniques. The merits of ITS and other two knowledge based learning experiences namely Domain Model and Learner model over traditional learning methods is discussed and the conclusion that can be attained is that the new method of teaching and learning is more effective and less time consuming than traditional method. Its demerit is that an ITS that is developed for a particular learning experience may not be suitable for another situation. If ITS is not timely updated, it will become outdated like traditional learning situation. Moreover, it is able to undertake educational tasks more than a human teacher and capable to interact with a learner to with a group of learners without any difference. So, ITS plays the very role of a virtual teacher and is able to innovate itself according to changing circumstances. References Aimeur, Esma., Frasson, Claude., & Dufort, Hugo. (2000). Cooperative learning strategies for intelligent tutoring systems. Informa world. 14(5), 465-489. Retrieved October 23, 2008, from http://www.informaworld.com/smpp/content~content=a713802506~db=all~order=page Cho, Byung In. (2000). Chapter 1: Problem statement: Dynamic planning models to support curriculam planning and multiple tutoring protocols in intelligent tutoring systems. Retrieved October 23, 2008, from http://www.cs.iit.edu/~circsim/documents/bcdiss.pdf Domeshek, Eric A., Knerr, Bruce W., & Howse, William R. (2008). Phase II final report on an intelligent tutoring system for teaching battlefield command reasoning skills. Storming Media: Pentagon Reports: Definitive: Complete. Retrieved October 23, 2008, from http://www.stormingmedia.us/07/0791/A079124.html Freedman, Reva. (2000). What is an intelligent tutoring system? Final Draft: Published in Intelligence. 11(3), 15-16. Retrieved October 23, 2008, from http://www.cs.niu.edu/~freedman/papers/link2000.pdf James Ong., & Soumya Ramachandran. (2003). Intelligent Tutoring Systems Using AI to Improve Training Performance and ROI: The intelligent tutoring system approach. Stottler Henke. 2. Retrieved October 23, 2008, from http://www.stottlerhenke.com/papers/ITS_using_AI_to_improve_training_performance_and_ROI.pdf Mills, Chris., & Dalgarno, Barney. (2007). The domain model: How to intelligent tutors represent and manage domain knowledge: A conceptual model for game based intelligent tutoring systems. Retrieved October 23, 2008, from http://209.85.175.104/search?q=cache:en7jugm4D9sJ:www.ascilite.org.au/conferences/singapore07/procs/mills.pdf+domain+model+of+intelligent+tutoring+system&hl=en&ct=clnk&cd=2&gl=in Rebecca S., et al. (2007). Intelligent tutoring systems: Evaluation of an intelligent tutoring system in pathology: Effects of external representation on performance gains,metacognition, and acceptance. Jamia: The journal of the American Medicasl Informatics Association. 14(2). 182-190. Retrieved October 23, 2008, from. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2213473 Reichgelt H, Major, N. (1990). Article information: Teaching strategies in the classroom. IEEE Xplore. Retrieved October 23, 2008, from. http://ieeexplore.ieee.org/Xplore/login.jsp?url=/iel3/1724/4573/00181200.pdf?arnumber=181200 Read More