This chapter will incorporate the reappraisal of literature of old research that is considered significance in the development of Intelligent Tutoring System for primary school pupils.
2.1 Problem Sphere
Problem sphere is the country that needs to be examined to work out a job. In this undertaking, Intelligent Tutoring System is used in the sphere of instruction. Education is the field for larning and learning. It is the procedure where cognition is transferred and received.
The intent to hold Intelligent Tutoring System for this undertaking is because to give an end product larning to the pupils particularly primary school pupils. Nowadays, instructors or pedagogues frequently face troubles to manage their pupils. It is because one instructor needs to provide many categories and each category will hold about 20 to 30 pupils. It is impossible to provide each pupil needs and penchant.
Each pupil have their ain acquisition manner either they are good in listening, visualising, or making stuff at manus. Since the instructor is impossible to cognize each pupil larning manner, hence, there are demands for Intelligent Tutoring System that can provides a tutoring system that can find the pupils involvement so they will non holding jobs such as deficiency of apprehension and misconceptions. Besides that, the benefit of this tutoring system towards the instructor is the instructor will easy supervise the pupil ‘s public presentation and they will cognize the suited attack to cover with the pupil manners of acquisition.
2.1.1 Learning Manners
Students learn in many ways such as by seeing and hearing, reflecting and moving, concluding logically and intuitively, memorising and besides visualising ( Felder, n.d. ) . Teaching and larning are different between each individual. It ‘s all depends on the persons itself. Everybody has larning manner strengths in which different people will hold different strengths ( Dunn, 1990 ) .
In 1986, Marie et al. , as cited in ( Farwell, 2000 ) provided an analysis in which approximately 20 to 30 per centum of the school-aged pupils remembers what is heard, 40 per centum easy recalled what they have seen or read and the remainder were normally used both techniques which is they heard and visualise at the same clip. They have their ain manner that will assist them in larning.
There are several different theories refering acquisition manners. Auditory, kinaesthetic and ocular are three types of cardinal acquisition manners ( Graham, n.d. ) . Below are the descriptions for each acquisition manners as cited in Graham ( n.d. ) and Farwall ( 2002 ) .
Childs who are audile learner normally prefer more on listening to account by reading them and sometimes they like to analyze by declaiming information aloud. Furthermore, audile scholar may love to environ with music while analyzing or they may necessitate a quiet infinite to analyze without diverted with any sounds. Auditory scholars learn successfully when the manner of giving information are being spoken and presented verbally.
“ Show me and I will understand ” is the keyword for ocular acquisition manner. It is a pattern to make new information by looking at something and visualise it. Normally, those people with this sort of larning manner can catch information presented in chart or graph, but they may foster impatiently listening to an account.
Most of the school ‘s kids excel through kinaesthetic which means touching, feeling, and sing the stuff at manus. Learning activity such as scientific discipline lab, field trip, skit and many other activities are the best technique for kinaesthetic scholar.
Most people use the combination of manner to acquire best acquisition manner for themselves. As for this undertaking, there are two larning manners covered which is the ocular and audile acquisition manners. This learning manners can be classify via some set of personality inquiries in which it will find the pupils country of acquisition manners.
Intelligent Tutoring System will normally come across to several techniques such as Case Based Reasoning technique, Agents technique, Neural techniques, Neuro Fuzzy techniques, Track Analysis and many more. Below are other potencies techniques that can be used for this Intelligent Tutoring System ‘s undertaking despite the Agents technique.
2.1.2 Examples of utilizing Bayesian Networks for Learning Styles Detection
Bayesian Networks is one technique that detects pupil ‘s acquisition manners in a web-based instruction system. In 2005, Garcia et al. , had proposed this technique to guarantee that all the pupils can larn even though they have different acquisition manners. Furthermore, Garcia et Al. ( 2005 ) , had besides stated that intelligent agent can used those information to gives the pupils personalized aid and present learning constituents that suit best harmonizing to pupil ‘s acquisition manners.
Table 2.1 shows the dimensions of the acquisition manners. Detectors like particulars informations and testing ; intuitive prefer political orientation and theories. Detectors are digesting with item but do non like complications ; intuitive are uninterested by item and love complications.
Table 2.1 Dimensions of Felder ‘s acquisition manners
( Beginning: Gracia et al. , ( 2005 ) )
A Bayesian Networks ( BN ) is a directed acyclic graph encodes the dependance relationships between a set of variables ( Pardalos, n.d. ) . It allows us to detect new cognition by uniting adept sphere cognition with statistical informations. In this BN, the nodes represent the different variables that determine a given acquisition manner. The arcs represent the relationships among the acquisition manner and the factor finding it. As shown in Figure 2.1, the theoretical account merely has the three dimensions of Felder ‘s model, perceptual experience, processing and apprehension.
Figure 2.1 Bayesian Network patterning pupil ‘s acquisition manners.
( Beginning: Gracia et al. , ( 2005 ) )
The undermentioned sentences describe in item the different states the independent variables can take:
Forum: station messages ; answers messages ; reads messages ; no engagement.
Chat: participates ; listens ; no engagement.
Mail: utilizations ; does non utilize.
Information entree: in tantrums and starts ; sequential.
Reading stuff: concrete ; abstract.
Exam Revision ( considered in relation to the clip assigned to the test ) ; less than 10 % ; between 10 and 20 % ; more than 20 % .
Exam Delivery Time ( considered in relation to the clip assigned to the test ) ; less than 50 % ; between 50 and 75 % ; more than 75 %
Exercises ( in relation to the sum of exercisings proposed ) : many ( more than 75 % ) ; few ( between 25 and 75 % ) ; none
Answer alterations ( in relation to the figure of inquiries or points in the test ) : many ( more than 50 % ) ; few ( between 20 and 50 % ) ; none.
Entree to illustrations ( in relation to the figure of illustrations proposed ) : many ( more than 75 % ) ; few ( between 25 % and 75 % ) ; none
Exam Consequences: high ( more than 7 in a 1-10 graduated table ) ; medium ( between 4 and 7 ) ; low ( below 4 ) .
The chance maps associated with the independent nodes are bit by bit obtained by detecting the pupil interaction with the system. 30 Computer Science Engineering pupils have been interviewed to find the values by experimentation utilizing the ILS ( Index of larning manners ) questionnaire. Then, allow the pupils used the instruction system and recorded their interactions with the system. The information was used to find the parametric quantity of the BN.
The Bayesian theoretical account is continuously updated as new information about the pupil ‘s interaction with the system is obtained. The chance maps are adjusted to demo the new observations or experiences. The chances reach equilibrium at certain point in the interaction. The chance values show a really little fluctuation as new information is entered. The values obtained at this point represent the pupil ‘s behaviour.
This paper considered for each dimensions three values to do the consequences more comparable. For illustration, for the understanding dimension, it considered the values consecutive, impersonal and planetary. The per centum of happenstances is 100 % for the understanding dimension, 80 % for the perceptual experience dimension, and 80 % for the processing dimension. All information from this paper is cited from Gracia et Al, ( 2005 ) .
In this undertaking, Intelligent Tutoring System is used to sort pupils larning manners. It used Numberss of regulations as the chief technique because it has the possible to give appropriate end product which is the acquisition manners for the pupils. Below are the descriptions of all techniques that will be used in this undertaking.
2.2.1 Intelligent Tutoring System
An early lineation of Intelligent Tutoring System ( ITS ) demands was delivered by Hartley and Sleeman in 1973 ( Shute & A ; Psotka, n.d ) . As stated by Shute and Psotka, Hartley and Sleeman argued that ITS must possess cognition of the sphere ( adept theoretical account ) , cognition of the scholar ( student theoretical account ) , and cognition of learning schemes ( coach ) . Furthermore, in order for ITS to hold appropriate control schemes, it need to hold capturing environment of acquisition, effectivity of communicating and to hold flexible determinations. The ITS is a plan in which pupil can communicated through a sequence of natural linguistic communication inquiries and replies and the coach could both ask and reply inquiries and maintain path of ongoing duologue construction ( Corbett, Koedinger & A ; Anderson, 1997 ) .
The authoritative ITS architecture consist of four constituents which are a undertaking environment, a sphere cognition faculty, a pupil theoretical account and pedagogical faculty.
Figure 2.2 ITS architecture
( Beginning: Corbett, Koedinger & A ; Anderson, ( 1997 ) )
As cited in Corbett, Koedinger & A ; Anderson ( 1997 ) , pupils engage in job resolution environments and these actions are evaluated with regard to the sphere cognition constituents. Student ‘s cognition province is maintained based on the rating theoretical account. Finally, the pedagogical faculty delivers instructional actions based on the rating of pupil ‘s actions and on the pupil theoretical account.
Advantages of ITS as described by Yousoof, Sapiyan & A ; Kamaludin ( 2002 ) , ITS is a systems that can supply considerable flexibleness in presentation of stuff and greater adaptability to react to idiosyncratic pupils need. It besides found to be extremely effectual in their intent. It has been proved by research that the pupils who tend to larn utilizing ITS really could larn fast when compared to the pupils utilizing traditional manner of instruction.
Disadvantages of ITS as besides cited in Yousoof, Sapiyan & A ; Kamaludin ( 2002 ) , hazard issues affects the execution of ITS, unsuccessful ITS can do the barrier in execution of ITS, replacing of human coach will besides be a barrier in execution and broad spread of ITS will take topographic point in another five old ages.
2.2.2 Rule Based Expert System
Expert system is a computing machine plan that uses cognition and illation process to work out job that are hard plenty to necessitate important human expert to work out for their solution ( Negnevitsky, 2002 ) . It is besides a computing machine plan in which it is able to execute at the degree of a human expert in a all right job country. The most popular expert systems is a regulation based expert system. It besides called as production regulations in which it contains IF-THEN statement.
Structure of Rule Based Expert System
A rule-based expert system has five constituents: the cognition base, the database, the illation engine, the account installations, and the user interface.
Figure 2.3 Basic Structure of Rule Based Expert System
( Beginning: Negnevitsky, ( 2002 ) )
The cognition base contains the sphere cognition utile for job resolution. In a rule-based expert system, the cognition is represented as a set of regulations. Each regulation specifies a relation, recommendation, directive, scheme or heuristic and has the IF ( status ) THEN ( action ) construction. When the status portion of a regulation is satisfied, the regulation is said to fire and the action portion is executed.
The database includes a set of facts used to fit aligned with the IF ( status ) parts of regulations stored in the cognition base.
The illation engine brings out the concluding whereby the expert system reaches a solution. It links the regulations given in the cognition base with the facts provided in the database.
The account installations enable the user to inquire the expert system how a peculiar decision is reached and why a specific fact is needed. An adept system must be able to explicate its logical thinking and warrant its advice, analysis or decision.
The user interface is the agencies of communicating between a user seeking a solution to the job and an expert system.
The user is the 1 who will be used the system. User is besides the 1 that will seek for solution.
Advantages of Rule Based Expert System
Natural cognition representation.
An expert normally explains the job work outing process with such looks as this: ‘in such-and-such state of affairs, I do so-and-so ‘ . These looks can be represented rather of course as IF-THEN production regulations.
two. Uniform Structure.
Production regulations have the unvarying IF-THEN construction. Each regulation is an independent piece of cognition. The really sentence structure of production regulations enables them to be self-documented.
three. Separation of cognition from its processing.
The construction of a rule-based expert system provides an effectual separation of the cognition base from the illation engine. This makes it possible to develop different applications utilizing the same expert system shell. It besides allows a graceful and easy enlargement of the expert system. To do the system smarter, a cognition applied scientist merely adds some regulations to the cognition base without step ining in the control construction.
Disadvantages of Rule Based Expert System
Opaque dealingss between regulations
Although the single production regulations tend to be comparatively simple and self-documented, their logical interactions within the big set of regulations may be opaque. Rule-based systems make it hard to detect how single regulations serve the overall scheme. This job is related to the deficiency of hierarchal cognition representation in regulation based expert systems.
two. Ineffective hunt scheme
The illation engine applies an thorough hunt through all the production regulations during each rhythm. Adept systems with a big set of regulations ( over 100 regulations ) can be slow, and therefore big rule-based systems can be unsuitable for real-time applications.
Inability to larn
In general, rule-based expert systems do non hold an ability to larn from the experience. Unlike a human expert, who knows when to ‘break the regulations ‘ , an expert system can non automatically modify its cognition base, or adjust bing regulations or add new 1s. The cognition applied scientist is still responsible for revising and keeping the system.
All information for Rule Based Expert System is cited from Negnevitsky ( 2002 ) .
2.2.3 Intelligent Agent
An agent is anything that can be viewed as comprehending its environment through detectors and moving upon that environment through effecters ( Rusell & A ; Norvig, 1995 ) . Presently, agents are the point of involvement on the portion of many countries of Computer Science and Artificial Intelligence.
Harmonizing to Jennings & A ; Wooldridge ( n.d. ) , an intelligent agent is a computing machine plan that is able to execute immediate response in order to run into its design aims. Flexible here means that the systems must be antiphonal in which agents should separate their environments and react in a timely to alterations that occur in it. Agents should besides be proactive whereby they should be able to exhibit chances, purposive behaviour, and take the enterprise where appropriate. Finally, agents should be societal in which agents should be interrelate when they comfortable with other Artificial Agents and worlds in order to finish their ain job resolution and to assist others with their activities.
Advantages of utilizing Intelligent Agent are because agents represent a powerful tool for doing system more flexible. Agents should act like an ‘expert helper ‘ with regard to some application, knowing about both the application and the user, and capable of moving with user in order to accomplish the user ‘s ends. Agents are besides good in bettering the efficiency of Software Development.
The restrictions or the disadvantages of utilizing agent as discussed by Jennings & A ; Wooldridge ( n.d. ) are: –
No overall system accountant
An agent-based solution may non be suited for spheres in which planetary restraints have to be maintained, domains where a real-time response must be guaranteed, or in spheres in which dead ends or unrecorded locks must be avoided.
No planetary position
Agents may do globally sub-optimal determinations since in about any realistic agent system ; complete planetary cognition is non a possibility. An agent ‘s action are by definition determined by that agent ‘s local province.
Trust and deputation
Users have to derive assurance in the agents that work on their behalf, and this procedure can take some clip. During this period of clip, the agent must strike a balance between continually seeking counsel ( and needlessly deflecting the user ) and ne’er seeking counsel ( an transcending its authorization ) . An agent must cognize its restrictions.
2.2.4 Multiagent System
As stated by Capuano et Al. ( n.d. ) , multiagent system ( MAS ) can be defined as loosely-coupled webs of pass oning and collaborating agents working together to work out jobs that are in front of their single capablenesss. In order to obtain consistent system behaviour, single agents in a multiagent system are non merely able to portion knowledge about jobs and solutions, but besides to ground about the procedures of coordination among other agents ( Capuano et al. , n.d. ) .
The thought of multiagent system is that an agent is a computing machine plan that has capableness to execute independent action on behalf of its proprietor or user. In add-on, agent can calculate out for itself what it needs to make in order to fulfill its design aims. A multiagent system is one that consists of figure of agents, which interact with another, typically by interchanging messages through some computing machine substructure ( Wooldridge, 2002 ) . In order to successfully interact, these agents will therefore necessitate the ability to collaborate, co-ordinate and negotiate with each other.
2.2.5 Distributed Case Based Reasoning
Case Based Reasoning ( CBR ) is another technique that is widely used in Intelligent Tutoring System and in the field of instruction so. As proposed by Rishi et Al. ( 2007 ) , they combine both technique which are CBR and agent technique to supply pupil patterning for online acquisition in a distributed environment with the aid of agents.
In this paper, it focused more on Case Based Distributed Student Modeling ( agent based ) ITS architecture to back up student-centred, self-paced, and extremely synergistic acquisition. The first measure is to construct the effectual acquisition environment which is the CBR where the system maintains a complete and full set of instances ( scenarios ) of pupil ‘s acquisition form and employs an efficient and flexible instance retrieval system.
The system as cited in Rishi et Al. ( 2007 ) must used the pupil ‘s larning profile such as larning manner and background cognition in selecting, forming and showing the larning stuff to back up instance based acquisition. As Rishi et Al. ( 2007 ) cited from Yi Shang et Al. ( 2001 ) and Kumar ( 2005 ) , Distributed CBR based pupil patterning enables adaptative bringing of educational contents and facilitates automatic rating of larning results.
This system consists of three agents with different expertness. The first agent which is personal agent will concentrate on pupil profiler which include cognition background, larning manner, involvements, class enrolled etc. The other two agents communicated with each other through different communicating channel which situated in distributed environments are learning agent and class agent. Figure 2.4 show the communicating theoretical account among agents.
Figure 2.4 Communication theoretical account among agents
( Beginning: Rishi et Al. ( 2007 ) )
Furthermore, the undermentioned activities as shown in figure 2.5 return topographic point during the pupil patterning when the pupil interacts with the system as such, choice of subject by the pupil and acquire to cognize pupil ‘s background by showing jobs to the pupil, analysing the pupil ‘s response by the system, choice of instance by the system based on response, version of the instance by the system, accomplishing the cognition constituent of the pupil theoretical account through instance retrieval, coevals of learning scheme by the system and showing the following job to the pupil.
Figure 2.5 Procedure of Student Modeling
( Beginning: Rishi et Al. ( 2007 ) )
Finally, this system is to the full distributed in which it does non bounded with any web topology, it reduces the demand of big storage infinites at the user ‘s site to hive away all the instances and redundancy is maintained for mistake tolerance. The whole system is managed in the distributed environment with merely three agents which are Personal Agent, Teaching Agent and Course Agent.
2.2.6 Path Analysis
Learning indexs is the manner of working and analysing the paths and supplying cognition on the activities ( Bousbia et al. , n.d. ) . This will assist instructors to comprehend and construe the scholar ‘s activities in e-learning state of affairss. As in figure 2.5 this paper by Bousbia et Al. ( n.d. ) considers three stairss in the analysis.
The first 1 is index ‘s pick. The first measure is fundamentally to steer the aggregation procedure. It helps the instructor to take high degree indexs in which the instructor intends to seek from the indexs base. It will so inquire the instructor to supply extra informations required for their computations. The following measure is observation. In this phase, the system identified the necessary paths extracted through a aggregation tool which installed on the learner side. This tool has specific history such as visited pages URLs, clip and actions. Finally, the analysis and interpretation measure. This is the most of import measure in which it divided into three chief phases which are shoping way rebuilding, indexs ‘ computation and learning manner tax write-off.
Figure 2.6 Learning Style Deduction Steps
( Beginning: Bousbia et Al. ( n.d. ) )
There will be three beds remain which are educational penchant bed, larning procedure bed and cognitive abilities bed. The first bed includes properties related to the preferable acquisition clip, environment penchant, information representations and encoding methods. The 2nd bed includes larning scheme, comprehension and patterned advance attack. For the last bed, it includes motive and concentration capacity.
The learning manner can be determined by ciphering the value of each bed ‘s property. By utilizing the necessary high degree indexs, the value is deduced. Furthermore, to link the indexs to the acquisition manners, Bousbia et Al. ( n.d. ) sort them harmonizing to theoretical account beds. The possible values of each bed ‘s property are chosen from the bing acquisition manner theoretical accounts, by doing their definitions closer.
2.3 Related Plants
Related plants are plants from other research workers which have related to this undertaking or possibly the same technique used but in different field or sphere. Intelligent Tutoring System and some other techniques is the chief focused in this research to compare and distinguish sphere and techniques with other undertakings.
2.3.1 Intelligent Agent in E-commerce
Ecommerce or e-commerce is the ability and accomplishment of selling merchandises or services over the Internet ( Ward, 2010 ) . As discussed by Pivk & A ; Gams ( n.d. ) in their article on Intelligent Agent in E-commerce, the article discussed on appraisal of agent engineerings which involved in purchasing and selling. Several agent-mediated electronic commercialism systems are analyzed in the position of a general theoretical account of the purchasing procedure.
E-commerce involves business-to-business ( B2B ) , business-to-customer ( B2C ) and customer-to-customer ( C2C ) minutess. It encounters a broad scope of issues such as security, trust, electronic merchandise, catalogues and many more. Intelligent agent can be used or applied to any of these.
Pivk & A ; Gams ( n.d. ) had given illustrations on the use of agent in ecommerce such as Tete-a-Tete ( T @ T ) . For illustrations in Figure 2.7, a shopping agent may have proposals from multiple gross revenues agents. Each proposal defines a complete merchandise offering including a merchandise constellation, monetary value and the merchandiser ‘s value-added services. The shopping agent evaluates and order these proposals based on how good they satisfy its proprietor ‘s penchants. If the shopper is non satisfied, he can review them along one or more dimensions. User shopping agent broadcasts this penchant changes to the gross revenues agents in which, in bend, utilize them to counter-propose better merchandise offering.
Figure 2.7 Consumer-owned shopping agents integrative negotiate with multiple merchant-owned gross revenues agents.
( Beginning: Pivk & A ; Gams ( n.d. ) )
2.3.2 Intelligent Agent Based Graphic User Interface ( GUI ) for e-Physician
This paper is proposed by Jung, Thapa & A ; Wang ( 2007 ) . It is all about the attack of utilizing ontology based intelligent interface agent that will help the doctor to get on-line entree interface to patient ‘s chart, fast rescheduling such as exigency instance, easy entree to research lab consequences and cut downing overall cost because of optimal use of clip.
In this paper, medical homecare system model is designed in real-time environment. There are four types of agents that are used in this system which are Interface Agent, Admin Agent, Laboratory Agent, Diagnosis Agent and Schedule Agent as in Figure 2.8.
Figure 2.8 Conceptual Framework of Intelligent Agent
Based user interface
( Beginning: Jung, Thapa & A ; Wang ( 2007 ) )
As cited from Jung, Thapa & A ; Wang ( 2007 ) , interface agent is the agent that will interact with the user and will work as an information filtering agent and choose the most critical instance per precedence. On the other manus, research lab agent will be able to supply the item scrutiny study from the research lab database. Furthermore, diagnosing agent will assist the interface agent to propose proper diagnosing by utilizing determination devising regulations. In add-on, administrative agent will supply pre-historic diagnosing tendency of the patient and eventually schedule agent will assist in fixing patient ‘s chart, programming, and fat rescheduling of the program on footing precedence. Those agents will assist in the development of the system and will give user ‘s concluding control for optimisation of their best Graphical user interface.
2.3.3 Intelligent Agent in Computer Games
Games are the practical universes that are more traceable than existent universe. It is besides something that can be controlled, formal, and mensurable, supply realistic and important challenge ( Mikkulainen, n.d. ) . Intelligent agents can be deployed in games today.
As cited in Lent et Al. ( n.d. ) , they discussed on the usage of intelligent agent in the games called Soar. It make the growing of intelligent agents for games easier by giving common illation engine and reclaimable cognition base that can be merely applied in many different games. Soar allows easy decomposition of the agent ‘s action through hierarchy of operation. It used Quake II and Descent 3 agents in which both have the functionality in the games such as winging in a starship without gravitation, onslaught, explore and many more.
Furthermore, Soar invariably cycles through perceive in which it accept sensor information from the game, think ( choice and execute relevant cognition ) and Act ( Execute internal and external actions ) . Interface is another of import portion in developing games since the interface extracts the necessary information from the game and encodes it into the format required in Soar.
2.3.4 Nervous Network-based Fuzzy Modeling of the Student in ITS
This paper is utilizing empirical attack that use the neuro-fuzzy synergy to measure the pupils in the context of an ITS is presented. Stathacopoulou, Magoulas & A ; Grigoriadou ( n.d. ) stated that fuzzed logic techniques is widely used in ITS since it have the ability to manage imprecise information such as pupil ‘s actions and to supply human descriptions of cognition and of pupil ‘s cognitive abilities.
In this paper, fuzzed logic is used to supply human-like approximative diagnosing of pupil ‘s cognition and cognitive abilities and Neural Network is used to trained human instructor ‘s determinations sing pupil ‘s features and fixed weight Neural Network are used to measure and aggregate rank map.
The neuro-fuzzy theoretical account has been tested in natural philosophies domain to measure pupil ‘s features for make up one’s minding about the appropriate instruction scheme. Experiments have been performed by Stathacopoulou, Magoulas & A ; Grigoriadou ( n.d. ) utilizing a population of 300 fake pupil instances with the determinations of 5 instructors. The overall mean categorization success has been 95 % . As decision, rating of pupils depends on interior decorator ‘s ability to analyse the cognitive sphere appropriately, define fuzzed sets and associate the pupil response with suited cognition and cognitive features.
2.3.5 FlexiTrainer: A Ocular Authoring Framework for Case-Based Intelligent Tutoring System
FlexiTrainer is an authoring model that allowed the fast fleet design of didactically rich and performance-oriented acquisition environments with tradition content and tutoring schemes ( Ramachandran, Remolina & A ; Fu ( n.d. ) ) . This authoring tool specifies a dynamic behaviour of tutoring agents that interact to present direction. FlexiTrainer has been used to develop an ITS for preparation chopper pilots in winging accomplishments.
FlexiTrainer consists of two constituents which are the authoring tools and the everyday engine. Core constituent for FlexiTrainer are Task-Skill-Principle Editor, Exercise Editor, Student Model Editor, and Tutor Behavior Editor. Each of these editors has their ain specific functionality. An instructional agent is used to transport out teaching-elated to accomplish instructional ends. It used Bayesian illation to integrate pupil patterning schemes.
2.3.6 Intelligent Tutoring System utilizing Hybrid Expert System with Speech Model in Neural Networks
This paper used supervised larning nervous webs to successful rate. Besides being more information bringing systems, this system aid pupils to actively build cognition. This paper by Venkatesh, Naganathan & A ; Maheswari ( 2010 ) enable learning system to be developed in assorted Fieldss and topics.
Nervous Model in this system is used for Question Answering System. As shown in Figure 2.9, input bed contains the inquiries on the wanted topics. Both possible inquiries and replies are stored in the coveted end product.
Figure 2.9: Nervous Network Architecture
( Beginning: Venkatesh, Naganathan & A ; Maheswari, ( 2010 ) )
On the other manus, address theoretical account consists of linguistic communication extraction ( includes categories such as noun, verb, operator, pronoun and many more ) , speech act classifier ( tutor uses strings of words and punctuation to sort each part of the scholar into speech act units ) , file direction ( used as marker for the lector ‘s manner reply which communicated with ITS faculty ) and manners ( choice individual to be communicated with the ITS either pupil, lector or admin ) . This system non merely cut down development times but besides appreciably simplifies the proficient cognition required of forces involved in the coevals of an auto-regulated intelligent tutoring duologue system ( Venkatesh, Naganathan & A ; Maheswari ( 2010 ) ) .
There are many ways in developing Intelligent Tutoring System as mentioned above. Each technique used has its ain strengths and failings. In this undertaking, Rule-based is used since it gives more impact and significance to the paradigm.
The following chapter will demo the research model on the methodological analysis for developing this paradigm.