Xuzhou Normal University Information Communication
Graduation thesis (design) opening report
The Application and Development of Intelligent Agent in Network Teaching Platform Design
Student number 01830417
Student Name Cheng Xiaoxia
Major education technology
Year level 01
Guiding teacher Yang Cheng
Fill in January 20, 2005
1. The purpose and significance of the source and research of the topic With the continuous development of computer technology, network communication technology, and multimedia technology, many changes have been made in teaching mode and teaching, and web-based instruction will be delivered. Network teaching is guided by the construction of educational psychology, through the Internet, using multimedia information processing technology, providing virtual classroom, conducting contextual teaching, promoting and supporting online learning. However, as far as the current technology is concerned, the requirements of education theory are far away; there are still many problems in the implementation of online teaching. For example, the teaching system is basically providing teaching materials, the material is rich but summate, or paying attention to the definitive details; the interaction is not enough, the student class discussion (such as BBS / QQ), because there is no on-site control mechanism, it is impossible to focus; Unable to teach different levels, the same teaching strategy is taken for different levels; the initiative of learning is mastered by students. It is prone to network value, deviation from learning purposes in the process of learning; the evaluation means is single, basically used to distinguish or error Fundamentally promotes the referee and self-expression of the learner. The above problems, except for subjective understandings and treatment, mainly due to technical reasons. Network teaching systems must be able to determine and adjust teaching strategies based on students' individual situations, with automatic monitoring and incentive functions, with intelligence. Intelligent Agents in Artificial Intelligence (AI) (Agent) can take action to achieve a scheduled goal by perceived self-learning in their own and environment. Therefore, an agent-based intelligent network teaching platform is introduced in modern educational technology, an agent-based intelligent network teaching platform is a key step in achieving intelligent network education. Since the mid-1990s, the application of intelligent agent (Agent) technology is increasingly common in web-based network teaching. The past smart auxiliary teaching (ICAI) mainly emphasizes the "teaching" of the teacher, using the "Student Model", "Knowledge Reasoning" and "Teaching Decision", help teachers achieve targeted teaching and targeted guidance (so ICAI is often ITS-intelligent guidance system). Since the 1990s, with the rapid spread of multimedia and Internet, a new learning theory and teaching theory - constructivism theory is increasingly popular. This kind of theory emphasizes that teachers 'guidance is student-centered, not only to play a teacher's guidance, but also fully reflect the role of students' cognitive subject. Therefore, intelligence in teaching should not only be reflected in the "teaching" of the auxiliary teacher, but should be embodied in the auxiliary student "learning". Intelligent Agent Technology is in this context, gradually replacing ICAI (or ITS) and has become a mainstream technology for intelligence in teaching. This is because the smart agent can not only make a teacher agent, but also as a student agent, but also becomes a multi-faceted agent in the student learning process, that is, Multiagents.
The following is the technical feature of the intelligent agent: Agent definitions are proposed by WoolDridge and Jennings in 1995, including two: weak definitions - theAgent is a computer system supported by the agent software, autonomy, social ability (sociability) , Reactivity and Activeness, strong definition - except for all features with weak definitions, agents also have some human unique knowledge, belief, desire Inteention, Coordinate, Cooperate and other factors. Agent mainly has the following characteristics: autonomous, Agent runs to make certain decisions for behavior and status, have some kind of control. Reactivity, Agent can sense changes in the operating environment and react in real time. Social, Agent has the ability to communicate and writing. Activity, Agent can accept inspiration information, actively approaching the target. The Agent architecture is divided into single Agent structure and multiple Agent structures. 2. Research status and analysis of domestic and foreign in this direction
At home and abroad, the teaching and application of intelligent agency has been studied. Some people in China have proposed a thought of learning tasks, and complete various submissions from each agent and corresponding learning members, each sub-task member communicates through KQML. Sharing the research results of each other. But in this mode, students in the process of solving the problem have not been guided by a humanized teacher at any time, and in the process of such collaboration completion tasks, the appropriate monitoring and guidance is always in the process of completing the task. necessary. Therefore, some scholars have proposed that in the intelligent information search of the Agent application, learner Agent initializes the interested point of interest by the user interface and the Knowledge Base management system, which points to the user model part of the stored knowledge base. : The information search Agent uses the internal library, scheduler, reactor, decision maker, read the user model in the knowledge base, determines the search task, start the search engine to search for a few websites or the entire Internet online search, and there are The scholar proposes an optimization algorithm for the teacher Agent in the large-scale network of Internet. Teachers and students have their own online agents, and there is a Facilitator scheduler between the two, including the teacher agent through the linear table and The hierarchical structure is distinguished. When students encounter a problem in a certain aspect during the learning process, he will propose this issue to the Internet network through its own student agent, and the Facilitator accepts the questions raised by the student agent, then find the best teacher agent that best solves such problems. One premise of this study is the first to establish an online agent for each teacher, and the search price across the Internet is obviously large. Foreign research in intelligent agents is much more mature than domestic, and has already put into practical models and systems. Microsoft Agent is one of the most familiar people. It first appeared in Office 95, mainly used for office assistance, Microsoft called "Office Assistant". In 1997, Microsoft launched the Agent version 1.5. From the interface, it completely separated from the window restrictions displayed by the agent in Office 95, becomes very imageive, speech synthesis, speech recognition and other technologies to the Agent. This way Agent is transformed into the smart interface that can be listened. Microsoft Agent is a very developed technology. At present, there are many software for Microsoft Agent, such as the anti-virus assistant in Jinshan Taxi, and some English learning software uses Agent technology. Another typical teaching agent research application is a teaching agent developed by USC (South University of South California) Adele (long-distance education agent Easy Edition, Agent For Distance Education-Light Edition). Adele is a successful application of intelligent agents. Currently, Adele is currently used in medical education courseware, professor trauma and tumor medical courses. This is an educational agent designed to adapt to the virtual education environment. Its architecture implements the most basic educational role of educational role: express knowledge, monitoring students, and providing feedback, exploration issues, and tips and answers, these roles are through a realistic agent that can support continuous and multi-mode interactions with students. People are done. The Adele architecture is capable of executing clients on the web browsing environment, and can also work with each other with virtual environments generated by other benchmaking tools. The most important part of the Adele structure is the agent and the reasoning engine. The agent is a Java program that can separately use the network-based JavaScript interface. But this agent can guarantee the independence and scalability of the platform because it is written in Java language. The reasoning engine completes all controls and decisions, the decision is based on the case after selecting the case, the case task illustration and initial state, and the current agent updated with the student completed the case process.
During actual operation, the agent and the reasoning engine in Adlele are downloaded to the client. Domestic research is mainly concentrated in collaboration between intelligent agent and agents, while intelligent agents act as an autonomous software entity is more than this. Foreign research pay more attention to intelligent agent reactions and monitoring capabilities, paying attention to communication between agents, and always establishing models for learning process for different knowledge as the basis of research. Get more monitoring capabilities of intelligent agents, and realize the teaching agents of the student should be the focus of our research 3. The main research content uses the network teaching platform of intelligent agents, students can communicate with teachers through dedicated remote channels, and communicate around teaching, and then form effective control mechanisms. With this platform, learners can go to the relevant resource site to find and collect information associated with the current learning content, and handle daily learning transactions. It can also facilitate teachers teaching and management, and ensure safe and reliable data, form a relatively closed teaching channel. All in all, students, teachers can easily exchange online, students can easily query online resources, students and students, teachers, teachers and students can quickly transfer information, information, etc., thus forming two " The main body "," the teacher, the student, reaches the effectiveness of the application of the application. The specific meaning of Multiagents in the online teaching platform is that it can be used as a teacher agent to achieve targeted help and counseling of learners; they can act as a student's learning partner as a student agent. Equal discussion, communication, sometimes in response to the blurred or one of the understanding of certain important concepts, deliberately put forward the difficult, provoking debate, to promote learners to think more and analyze; A student agent acts as a visual friend, so that the learner encounters uncomfortable things or mentally obstacles to knee, helping to reduce the burden and eliminate psychological stress; in addition, can be used as a secretary agent to help learn To the relevant resource site to find and collect information related to the current learning content, or help learners handle daily transactions (such as collection, email, arrangement of appointment, prompt should be handed, help review, preparation ... etc.) . In summary, Multiagents is very useful in the process of teaching or learning, compared with ICAI (or ITS), can play more flexible and more sampled in intelligence. Based on Multiagents, follow the JXTA specifications, use limited network resources, make teachers, can communicate directly with students in the network teaching platform, and conduct counseling in learning, and can easily transfer documents in the network. , Make the teacher's full interaction.
Here are the points to describe the relevant technical points. Today, P2P technology is still in the initial development phase, P2P is a point-to-point (or partner partner) technology, abandoning the Internet situation of the past customers / services, making information exchange more convenient and fast. As a new generation of Internet communication, P2P is still in a blank in remote education, waiting to be developed. The JXTA technical specification is to make up for the defects of P2P technology. Based on this protocol, users can connect, do not need to use the P2P platform developed by them. JXTA technology provides basic mechanisms to solve problems facing current distribution computing applications, achieving new generations of unified, security, interoperability, and heterogeneous applications. It currently supports Java technology-based platforms and systems. The future JATX technology will support more small mobile devices without the limitations of memory. JXTA provides powerful functions through Java technology and XML data, providing vertical applications interacting, and can overcome restrictions in the current P2P software. At the same time, through small, simple, easy-to-develop construction modules, JXTA will enable developers to be liberated from the complex work of establishing their own frameworks, and can focus on building all kinds of novel, creative, distributed computing applications. The Java and JXTA specifications are open to the world software developers, in order to get rid of foreign software (especially operating systems, such as Microsoft window series) monopoly and constraints, the Chinese government actively encourages free operating systems such as Linux. And the procedures developed by Java don't have to worry about having copyright issues (because it is open), you can run on multi-platform (including Linux, Solaris and Windows Series), JXTA as a new generation of P2P technology solutions, Suitable for development specifications for this topic. 4. Research programs and schedules, the expected objectives intended to be subject to research methods and technical routes (including the overall arrangements, steps and progress, etc.): This topic is based on Java language, JXTA and P2P are technical specifications, and develop online teaching. platform. It is proposed to take the following way:
Stage 1: Analysis stage, research, and listen to teachers and students' needs. January 2005 to February 2005
Stage 2: Design phase, through research report, design flow chart. February 2005 to March 2005
Stage 3: Development procedures, according to the design phase process, encoding. March 2005 to April 2005
Stage 4: Test Maintenance Procedure, perform practical tests according to the actual situation. April 2005 to May 2005
Phase 5: Configure the server. Implement product input and complete graduation design tasks.
The 1st to 3 stages are not clearly defined, screw development
The results of the study and final results:
Staged results:
By January 2005, submit a feasibility investigation report;
By March 2005, the design of the OOD flow chart was completed;
By April 2005, the coding is completed;
By May 2005, the software test was completed, and the final scientific research results were submitted.
5. Conditions for completing the conditions and funds required to complete the subject: Development Language Java, Tools and Specification JXTA Requirements: Test Environment, eventually need to be disclosed on the hospital. It is expected that the difficulties and problems that may encounter during the study, and the measured measures 1) The theoretical problem intelligent agent technology is in the network teaching platform, and there is already in-depth research at home and abroad, but uses new technologies to study intelligent agents, and apply There are still less in the network teaching platform. Solution: Based on the theoretical basis, refer to the results of the survey of teachers and students, make appropriate theoretical analysis. 2) Technical issues Due to J2EE, JXTA is a technology that has emerged in recent years. I am also a first-time contact solution: Drilling research J2EE, JXTA technology principle, and other technicians, based on my theory, jointly developing the corresponding platform. 7. Main references [1] Robert Flenner Michael Abbott Toufic Boubez Frank Cohen Navaneeth Krishnan Alan Moffet Rajam Ramamuriti Bilal Siddiqui Frank Sommers with kaolinite LIU ZHOU Zhao indeed translate Java P2P technology insider Java P2P UNLEASHED People Post Press, 2003 (10) [2] Liu Da, et al. The current situation and development trend of the study [J] .Journal of Software, 2000 (10) [3] Microsoft Agent Series Microsoft Web Site http://www.microsoft.com [4] Master's thesis Based on the Internet based Teaching Support System Research 2003.5.1 [5] Wu Qiang Master's Paper Web Teaching Platform 2002.5.1 [6] Li Jie Masters The Fifth Media Agent Design 2003.3.1 [7] Zhu Suyuan Master's Papers Network Education Intelligence Research on Search Engine 2004.6.1 [8] Dong Dong Master's Thesis Based on MAS Tongji University E-Learning Platform Research and Design 2004.2.19
