Software Agents for Managing Learning Plans
Software Agents for Managing Learning Plans
复制标题
用于管理学习计划的软件代理
DOI：
10.28945/2994
复制
发表时间：
2006
期刊：
影响因子：
--
通讯作者：
I. Hawryszkiewycz
中科院分区：
文献类型：
--
作者：
I. Hawryszkiewycz
1.对更加个性化的学习的需求将需要新的方法来支持学习过程。这种需求可能是获得特定的能力（Eschemans和Ritzen，2002）或新学科领域的知识。教育方法也更加强调建构主义学习方法（Jonassen，2002年），学习者以自己喜欢的方式构建知识。学习者设定他们的学习目标，制定学习计划，并提供学习材料。在教育机构中，学习计划通常是一系列讲座和评估，所有学生都要参加一个科目。我们解决的方法来定制这样的计划，以专门的学习者需要大量的学习者。然后，使用支持系统指导学习者完成他们的专门计划，以尽量减少教师管理计划所需的工作。这种指导可以采取多种形式。它可能是为了确定缺乏一些基本知识，并提供在继续主要学习计划之前熟悉这些知识的方法。制定计划的另一个好处是支持建构主义学习，一些作者（Petraglia，1998）认为在当前的系统中没有有效的支持。个性化学习需要新型的支持系统。使用目前的技术将是非常昂贵的。一个老师显然是有限的学生数量，每个人都遵循他们可以支持的个人计划。代理系统可以在这里提供更好的支持，因为它们可以在学习者利用学习管理系统提供的服务的方式上提供更大的灵活性。代理人系统已经在早些时候提出。一种是教学代理人（Baylor，2003），如专家，激励者或导师。另一种是支持特定功能的代理（McArdle，2005），例如接口或导航代理。本文提出了第三类代理，即过程代理。过程代理引导用户完成过程，并且在它们独立于域的意义上是通用的。它们不包含领域知识，而只是帮助学习者和教师进行学习过程。他们帮助学习者获得材料，并在必要时创建他们的学习计划。早先已确定了两类加工剂（Hawryszkiewycz，2005年a）。一种是帮助学习者构建学习计划的代理。另一个是管理学习计划并根据需要动态修改计划的代理。早期的工作已经描述了代理，帮助学习者创建自己的学习计划。本文集中在管理学习计划的代理。定义过程我们定义通用代理的第一步是定义一个分类法或语法，用于描述学习过程。语法或分类法将提供一组基本的概念，作为通用代理系统的框架。图1中示出了用于学习过程的这种语法的主要元素或学习过程概念。这些包括：学习环境，或学习发生的地方。这可能是一所大学的一个人的工作场所。学习目标，它描述了学习目标，学习计划，它定义了实现学习目标所需遵循的学习活动顺序。学习活动，它描述了学习计划的一个步骤;这可能是创建一个报告，评估一个问题，主题元数据，它提供了明确的参考活动中所需的信息，学习方法，这将在学习步骤中使用，为学习方法提供支持服务。[图1省略]这里的一般语义是学习者指定学习目标。一个计划，这是由一些学习活动，然后由学习者从代理的帮助下构建。…
Introduction Demands for more customized learning will require new ways to support learning processes. Such demands may be to acquire particular competencies (Hezemans and Ritzen, 2002) or knowledge of new subject areas. Educational approaches are also placing more emphasis on constructivist learning approaches (Jonassen, 2002), where learners construct knowledge in their preferred ways. A learner sets their learning goal and develops a learning plan and is provided with learning materials. In educational institutions, the learning plan is usually a set of lectures and assessments that is followed by all the students taking a subject. We address ways to customize such plans to specialized learner needs for large numbers of learners. The learners are then guided through their specialized plan using support systems to minimize the effort needed to manage the plans by teachers. Such guidance can take many forms. It may be to identify a lack of some elementary knowledge and provide ways to build familiarity of this knowledge before continuing with the main learning plan. One further advantage in formulating plans is support constructivist learning, which some writers (Petraglia, 1998) suggest are not effectively supported in current systems. New kinds of support systems are needed for personalized learning. It will be prohibitively expensive to use current techniques. One teacher obviously is limited in the number of students, each following their personal program that they can support. Agent systems can provide better support here as they can provide greater flexibility in the way learners utilize services provided by learning management systems. Agent systems have been suggested earlier. One are the pedagogical agents (Baylor, 2003) such as expert, motivator or mentor. Another are agents that support specific functions (McArdle, 2005), such as interface or navigational agents. The paper proposes a third class of agents, namely process agents. Process agents guide users through a process and are generic in the sense that they are domain independent. They do not contain domain knowledge but simply assist learners and teachers to proceed through the learning process. They assist learners to get access to materials, and where necessary create their learning plans. Two classes of process agents have been identified earlier (Hawryszkiewycz, 2005a). One are agents that assist a learner to construct a learning plan. The other are agents that manage the learning plan and dynamically amend the plan as needed. Earlier work has described agents that assist learners to create their learning plans. The paper concentrates on agents that manage learning plans. Defining the Process Our first step in defining generic agents is to define a taxonomy, or grammar, for describing leaning processes. The grammar or taxonomy will provide a fundamental set of concepts that serve as the framework for generic agent systems. The main elements, or learning process concepts, of such a grammar for the learning process are shown in Figure 1. These include: Learning environment, or where learning takes place. This may be a University of a person's place of work. Learning goal, which describes the learning objective, Learning plan, which defines the sequence of learning activities to be followed to achieve the learning goal. Learning activity, which describes a step of the learning plan; this may be create a report, evaluate a problem, Subject metadata, which provides explicit references to information needed in the activity, The learning method, which will be used in the learning step, Support services provided for the learning method. [FIGURE 1 OMITTED] The general semantic here is that a learner specifies a learning goal. A plan, which is made up of a number of learning activities, is then constructed by the learner with assistance from an agent. …