結合VanHiele認知模式與貝氏網路之智慧型程式語言學習系統AnintelligentprogramminglanguagelearningsystembasedonHieleVanc.ppt

《結合VanHiele認知模式與貝氏網路之智慧型程式語言學習系統AnintelligentprogramminglanguagelearningsystembasedonHieleVanc.ppt》由会员分享，可在线阅读，更多相关《結合VanHiele認知模式與貝氏網路之智慧型程式語言學習系統AnintelligentprogramminglanguagelearningsystembasedonHieleVanc.ppt（27页珍藏版）》请在三一办公上搜索。

1、IVHPS:A Web-based Bayesian van Hiele Problem Solver for Java Language,J.Wey Chen,ProfessorDepartment of Information ManagementSouthern Taiwan UniversityTainan,Taiwan,Outline,Introduction*Motivation*Purpose of the study*Advantages of the SystemTheoretical Foundation*Van Hiele Model*The Cognitive Theo

2、ry*Bayesian network(BN)*General architectureDignostic test Results and DiscussionConclusion,2,On“Programming Teaching and Learning”,1.Programming is a complicated business.2.Dijkstra1 argues that learning to program is a slow and gradual process of transforming the novel into the familiar.3.programm

3、ing is not a simple set of discrete skills;the skills form a hierarchy,and a programmer will be using many of them at any point in time.,3,Purpose of the Study,This paper formulates an alternative pedagogical approach that encompasses the van Hiele Model,cognitive model,and Bayesian network to desig

4、n a web-based intelligent van Hiele Problem Solver(IVHPS).,4,Advantages of the System,The system takes full advantage of Bayesian networks(BNs)to:1.provide intelligent navigation support,and 2.make individualized diagnosis of student solutions in learning computer programming languages.,5,Theoretica

5、l Foundation,6,Van Hiele Model,7,Level 0Visualization,Level 1Analysis,Level 2Informal Deduction,Level 3Deduction,Level 4Rigor,The Cognitive Theory,8,Bonar and Soloway11 represented and arranged programming knowledge according to its level of difficulty in four cognitive levels:Lexical and Syntactic

6、Semantic Schematic Conceptual,The Combined Model,9,Knowledge structure for each learning node,Bayesian network(BN),10,A Bayesian network(BN)consists of directed acyclic graphs(DAG)and a corresponding set of conditional probability distributions(CPDs).Based on the probabilistic conditional independen

7、cies encoded in the DAG,the product of the CPDs is a joint probability distribution.,Using Bayesian Networks in Diagnostic Test,11,E,12,Chens Implementation(2006),13,Level 1Visualization,Level 2Analysis,Level 3Informal Deduction,Level 4Deduction,Level 5Rigor,Level 1Visualization,Level 2Descriptive&R

8、elations,Level 3Implications,Level 4Logic Modification&Analogy,Level 5Abstraction&Modeling,General architecture of intelligent van Hiele Problem Solver,14,A screen shot of IVHPS displaying the diagnostic report,15,A screen shot of IVHPS displaying the lecture notes for the concept“Data Types”,16,A s

9、creen shot of IVHPS displaying a typical quick-run sample output,17,A screen shot of IVHPS displaying a typical practice sample from the expert template,18,Dignostic test Results and Discussion,19,Knowledge Structure for Dignostic Test,20,-To move around the levels in a node,Discussion,21,Discussion

10、,To move to different learning nodes,22,Discussion,To determine the learning sequence,23,N4L3,N5L0,N6L0,N4L3,？,？,N6L0,N5L0,Discussion,Diagnosis,24,N5L3,N7L3,N8L0,？,N7L3,Conclusions,25,The success of this model is attributed to the extensive review of the available literature and to the exploratory i

11、nterviews with students who participated in the first phase of study.The proposed Modified van Hiele Model for Computer Science Teaching can help unveil the mystery of the“hidden mind”and provide a logical link for students to inductively learn problem-solving and programming skills.The system is ab

12、le to utilize Bayesian network techniques in modeling the student knowledge based on the proposed knowledge structure.,A Practical Model for Applications,26,To help engineering educators wisely utilize the information described in this paper,we suggest the following approach be taken to design sound

13、 curriculum content and sequence:Hold an expert roundtable discussion to roughly determine a set of knowledge concepts required for a course.Manually construct the course DAG with the aid of the course textbook.Develop a diagnostic test to have test questions which cover every cognitive category for

14、 every level of understanding in the entire curriculum structure.4.Extensively conduct the test and collect sufficient Bayesian training data.Analyze and use the Bayesian training data to trim the unrelated content and adjust the logical sequence for learning.Once the process is completed,a new course DAG will be produced.Group the related knowledge concepts into chapters according to their sequences appearing on the course DAG.,Thank you for your attention!,27,