Nowadays, there are more and more researches focused on prediction of learning outcome, and most of them applied quantitate type of analysis approaches. Thus, we want to apply another type of analysis approach to do early prediction. In this research, we applied temporal features and analysis approach to predict students' learning outcomes and identify at-risk students. The result shows that using temporal features is effective on early prediction of learning outcome and there exists differences of learning behaviors between students which have different learning background.

Many previous studies have shown that students' learning strategies and engagement are both important factors affecting students' learning outcome. We want to know whether the engagement of students who use different learning strategies is different, so that we can proxy measure students' engagement according to the learning strategies used by students. Different from the past methods of using questionnaires or interviews, this research used education data mining technology to directly analyze the learning logs of online learning system to extract students' learning strategies. We collected the learning logs of 56 students, and extracted the learning strategies of the students using sequence clustering. We classified four learning strategies and performed Pearson correlation analysis with students' learning outcome and engagement, and discussed the key strategies that affect students' learning outcome. We also found that students who use different learning strategies have different levels of engagement.

In recent years, educational resources have gradually been digitized, and digital education platforms have gradually become popular. We use AI to accurately assist people in performing daily tasks through a machine learning process. In education, we can use AI in many situations, such as predicting student's learning outcome and discovering student's learning strategies. However, most solutions have not yet utilized modern AI capabilities, such as natural language processing. This research aims to help teachers use machines to automatically generate short answer questions to reduce the time for teachers to write exam questions. In addition, the main reason we focus on short answers is that many studies prove that short answer exercises can enhance student's long-term memory, thereby improving their learning performance. We propose an automatic question generation (AQG) system that combines syntax-base and semantics-base, in order to prove that the system is highly available and improve student's learning performance, we conducted experiments with 41 students. The experimental results show that student's learning performance has been significantly improved, which means that by repeatedly testing the machine question generation system, students can deepen their long-term memory of course knowledge.

Elementary school students' mathematical ability plays an important role in the development of their future, but the proportion of low-achieving elementary students in mathematics in Taiwan is high due to their aversion to mathematics. The aim of this study is to establish a learning system to help low-achieving students' learning in mathematics. A 3D game-based learning system was developed in this study to achieve the aim and the students' learning outcomes were measured by the pre, post-test and their learning logs in the system. The participants were 6 elementary school low-achieving students studying in Northern Taiwan. The results showed that the system improved their learning outcomes. The design of the system can also be a reference for developing future virtual learning systems for mathematics.

Computer programming is an important competence for engineering and computer science students. Teaching and learning programming concepts and skills have been recognized as being a big challenge to both teachers and students. Accordingly, researchers have indicated that 'learning strategy,' 'lack of study' and 'lack of practice' are the causal attributes of success or failure in a computer programming course. Therefore, to cope with this problem, an interactive test system is proposed to enhance students' learning outcomes in computer programming courses in this study. An experiment has been conducted on a college computer programming course to evaluate the effectiveness of the proposed method. The experimental results show that the proposed approach can benefit the students in enhancing their programming skills. In addition, the students' attitudes toward learning computer programming were improved.

This study aimed to investigate effects of unidirectional versus reciprocal teaching strategies on computer programming learning. Thus, this study conducted an experiment with participating novice programmers, who applied no strategies, unidirectional, and reciprocal teaching strategies for computer programming learning by using a web-based learning system (VPen). Yet, this study explored participants' perceptions toward the VPen system and the strategies. The paper reports results of this study as well as it provides useful suggestions on application of the strategies by using the VPen for teaching and research community of this field and for future studies.

The purpose of this thesis is to implement a 3D situational English learning process which is based on Task-based approach. To achieve this goal, we created situational setting on Second Life and combined the framework of Task-Based Learning approach proposed by Jane Willis in 1996 to design a three-stage teaching process. Under the topic of daily basis English, we have designed an online 3D virtual English program for students to carry out situational English learning in the future.

Authentic context and active social interaction are two critical factors in the success of foreign language learning. Second life (SL) has been viewed as an attractive platform with potential for foreign language learning because of its virtual authentic real context. However, simply moving students to SL guarantee neither students’ active involvement and rich interpersonal interaction nor effective learning outcome. In a two-month pilot study of Chinese learning and teaching in SL, analytical results indicate that there is still more needed to do in SL to help create a more active Chinese language learning environment for Chinese learning and teaching for global learners. While this pilot study focuses on teaching and learning Chinese as a foreign language, the findings and suggestions are also relevant to understanding the related issues of other language.

This paper presents our design and development of a context-driven content adaptation planner, which dynamically transforms requested Web content into a proper format conforming to receiving contexts (e.g., access condition, network connection, and receiving device). Aiming to establish a semantic foundation for content adaptation, we apply description logics (DLs) to formally define context profiles and requirements and automate content adaptation decision. In addition, the computational overhead caused by content adaptation can be moderately decreased through the reduction of the size of adapted content.

This paper presents content adaptation technique for providing adaptive content delivery and presentation based on students' learning environment. We have applied content adaptation to enhance mobile learning on blackboard learning system in our university. The experiment results show that content adaptation can extend the time and space of learning and improve the collaboration among students.

Knowledge sharing in communities of practice requires intensive social interactions among community participants, such as articulating knowledge into written forms, posting messages to community's discussion forum, responding comments to the messages posted by others; namely annotation. Annotation is an explicit expression of knowledge in the form of comments which reveal conceptual meanings of annotators' implicit thought. This paper presents a personalized annotation management system (PAMS) and demonstrates how PAMS enhances knowledge sharing in communities of practice. An experiment has been conducted to evaluate how PAMS enhances knowledge sharing in students' online group reading activity. The result of the experiment has demonstrated that PAMS-enabled knowledge sharing do improve students' reading comprehension.

In the mobile Internet, users mostly work with handheld devices with limited computing power and small screens. Their access conditions also change more frequently. In this paper, we present a novel service supporting intelligent content adaptation to better suit handheld devices. The underlying technique is a unit of information (UOI)-based content adaptation method, which automatically detects semantic relationships among comprising components in Web contents, and then reorganizes page layout to fit handheld devices based on identified Ious. Experimental results demonstrate that our method enables more exquisite content adaptation.

Web service technology enables seamless integration of different software to fulfill dynamic business demands in a platform-neutral fashion. However, the adoption of loosely-coupled and distributed services will cause trustworthiness problems. In this article, we present an experience-based evaluation of service’s trustworthiness based on trust experience (understanding) and trust requirements (policy). We utilize ontology to specify past experiences of services and trustworthy requirements of the requester. Before invoking found services, the addressed method can help the requester evaluate the trustworthiness of the services based on his or her trustworthy requirements and past experiences of the services. Furthermore, we also present an evaluation method for composite services by taking the structure of the composite services into account. The main contribution of the paper is providing evaluation methods for Web services such that the service requester can make better decisions in selecting found services in terms of service’s trustworthiness.

Blended learning (BL) is a hybrid of classroom learning and online learning that includes the conveniences of online courses without the complete loss of face-to-face (F2F) contact. Advancements in communication technologies now allow us to have synchronous distributed interactions that occur in realtime with close to the same levels of fidelity as in the F2F environment. Therefore, it is imperative that we understand how to create effective BL experiences that incorporate both traditional instruction and online instruction. This article uses mathematics model to formulate the transactional distance to compare the interaction distance of traditional centralized F2F and synchronous distributed F2F. Transactional distance is then used to know about the communication gap between learner and instructor, content, and other learners. Finally, the paper verifies the validity of the transactional distance calculated through the mathematical model with the help of survey that elicits students’ perception towards those two environments.

In this paper, we apply peer-to-peer and instant messenger for the augmentation of knowledge sharing in collaboration supported virtual learning communities. We enhance current peer-to-peer and instant messenger with content quality control and social network to further improve content discovery and group discussion. Content found by peer-to-peer is referred to accessible explicit knowledge, while the discussion and interaction details recorded in instant messenger is referred to tacit knowledge. Results of this study demonstrate that with our enhanced mechanism, it can help learners find relevant content as well as knowledgeable collaborators who are willing to share their knowledge

Context aware learning service is an interactive model between learners and learning services based on learners' surrounding learning environments. We envision that providing context aware learning services is an important step to enhance current Web and mobile based e-learning by finding right collaborator, right information and right learning services in the right place at the right time. The major contributions of this paper are the provision of context-aware learning services, namely context-aware content access for learners with portable devices

Universal access is known as the concept of facilitating accesses to the desired data according to the varying capabilities and characteristics of users. Universal access aims to be capable of recognizing and responding to the presence of different context in a seamless, unobtrusive and often invisible way. To achieve this goal, two important issues need to be solved, they are context awareness and content adaptation. In this paper, context is referred to as contextual information that can be used to characterize the situation of an entity where an entity can be a person, a place, a physical or computational object. The contextual information can be where and when the users are (time and space), as well as what the content are available nearby, and etc

The objective of this paper is to present a new learning design framework to enhance learning resources reusability through content discovery, revision, and verification. Our learning design framework can help instructors to design their instructional/learning resources by reusing existing learning design documents and components rather than starting from scratch. Our framework provides a learning design finder in order to help instructors to locate learning design documents matching their instruction needs. If an exactly matched learning design document can be found, then the instructors can directly reuse the matched learning design document. If only partially matched learning design document can be found, then our framework provides a learning design composer to guide instructors to compose a new learning design document to fit instructors' needs. The newly found and modified learning design documents will be verified based on IMS learning design speciation rules before they can be fully reused.

The activities in learning community include posting easy questions or complex problems, sharing experiences and knowledge and pedagogical activities. Learners usually post questions when they have problems understanding some concepts. Information is usually accumulated at a rapid pace in learning communities. However, if information is not organized orderly and systematically, it will be difficult for community members and Webmasters to find relevant information to browse, to respond to, and to learn. This study proposes the semantic retrieving concept model (SRCM) which improves the efficiency of searching and learning tasks. When learners need information of certain topics, they simply enter conceptual keyword in the semantic conceptual search textbox. The content-based retrieval mechanism of SRCM will then retrieve the relevant discussion messages and e-contents from the content repositories.

In this paper, we propose a Petri nets formalism for the verification of rule-based systems. Typical structural errors in a rule-based system are redundancy, inconsistency, incompleteness, and circularity. Since our verification is based on Petri nets and their incidence matrix, we need to transform rules into a Petri nets first, then derive an incidence matrix from the net. In order to let fuzzy rule-based systems detect above the structural errors, we are presenting a Petri-nets-based mechanism. This mechanism consists of three phases: rule normalization, rules transformation, and rule verification. Rules will be first normalized into Horn clauses, then transform the normalized rules into a high-level Petri net, and finally we verify these normalized rules. In addition, we are presenting our approach to simulate the truth conditions which still hold after a transition firing and negation in Petri nets for rule base modeling. In this paper, we refer to fuzzy rules as the rules with certainty factors, the degree of truth is computed in an algebraic form based on state equation which can be implemented in matrix computation in Petri nets. Therefore, the fuzzy reasoning problems can be transformed as the liner equation problems that can be solved in parallel. We have implemented a Petri nets tool to realize the mechanism presented fuzzy rules in this paper.

Many mobile and wireless devices are connecting to the Internet nowadays, among them, mobile phones and PDAs are the most popular ones. Thus, in this paper, we will focus on how to develop embedded software running on the two devices by using design patterns and Java based software components. Notably, some components may be used directly in an embedded software system, whereas most components must be specialized prior to reuse. Developers have to identify the variation points on these components. In order to develop a reusable embedded software system, we will identify several variation points with some variants, and introduce some useful design patterns for implementing variation points. Consequently, we can customize an embedded software system just through attaching variants on corresponding variation point in our system. The design patterns we are using are property container, strategy, decorator, and model-view-controller. The component technology we are using is J2ME. J2ME is a specification focused on the development of mobile applications. It provides a similar environment as standard Java environment. J2ME's components programming includes the Spotlet programming for PDA, and the MIDlet programming for Java phone. In addition, along with the XML, J2ME can also provide the XMIDlet programming for dynamic downloading and execution of XML-based applications for PDA and Java phones.

The focus of the paper is on the use of time Petri nets to serve as the verification mechanism for acquired scenarios. Use cases are used to elicit the user needs and to derive the scenarios. After specifying all possible scenarios, each of them can be transformed into its corresponding time Petri nets model (TPN). Through the analysis of these TPN models, wrong information and missing information in scenarios can be detected. The proposed approach is illustrated by a course registration problem domain.

In the past several years, various graphical techniques were proposed to analyze various types of structural errors, including inconsistency (conflict rules), incompleteness (missing rules), redundancy (redundant rules), and circularity (circular depending rules), of rule based systems. We present a special reachability graph technique based on /spl omega/-nets (a special type of low-level Petri net) to detect all of the above types of structural errors. Our new technique is simple, efficient, and can be easily automated. We highlight the unique features of this new approach and demonstrate its application through an example.

In this paper, we propose an approach to reusing requirements specification, called task-based specifications in conceptual graphs (TBCG). In TBCG, task-based specification methodology is used to serve as the mechanism to structure the knowledge captured in conceptual models, and conceptual graphs are adopted as the formalism to express requirements specification. TBCG provides several mechanisms to facilitate the reuse of formal specifications: a contextual retrieval mechanism to support context-sensitive specifications retrieval and incremental context acquisition, a graph matching mechanism to compute the similarity between two graphs based on the semantic match and fuzzy logic, and a paraphraser to serve as an explanation mechanism for the retrieval specifications. ©1999 John Wiley & Sons, Inc.

This paper presents our reachability tree logic (RTL) and its integration with time Petri nets to specify and verify the temporal behavior of high assurance systems. The specification phase begins with a system modeling to model system requirements into a time Petri net N and construct a reachability tree RT of N. We then use RTL to specify the desired temporal behavior as formula F. The verification phase uses a model-checking algorithm to check whether RT can satisfy F, that is to find firing sequences to satisfy F. If F is not satisfied, we then modify N into N′ and obtain a RT′ of the modified N′. The modification (refinement) continues until the modified RT′ can satisfy F. In addition, we will demonstrate how to reduce the complexity of model-checking by using our RTL-based algorithm. We have implemented a specification and verification toolkit called NCUPN (National Central University Petri Nets toolkit) using Java. NCUPN is now available on the Internet via

We propose a Petri net formalism for the verification of rule based systems. Typical structural errors in a rule based system are redundancy, inconsistency, incompleteness, and circularity. Since our verification is based on Petri nets and their incidence matrix, we need to transform rules into a Petri net first, then derive an incidence matrix from the net. In order to allow a rule based system to be immune from the above described structural errors, we have observed that for all columns in the matrix, all positive entries must be above all negative entries; and for all rows in the matrix, all positive entries must be to the right of all negative entries. If this is not the case, the rule based system may commit errors. Based on this concept, we have developed a tool consisting of the following four phases: rule normalization, rule ordering, rule-to-Petri-net transformation, and rule verification. In phase one, we normalize the rules into Horn clauses. We rearrange the ordering of these normalized rules in phase two, then transform the reordered rules into a Petri net and its corresponding incidence matrix in phase three. In phase four, we perform the rule verification based on the incidence matrix.

A framework of integrated expert systems based on fuzzy Petri net, called fuzzy Petri net based expert system (FPNES) are proposed for bridge damage assessment. Major features of FPNES include: reasoning for uncertain and imprecise information; knowledge representation through the use of hierarchical fuzzy Petri nets; reasoning mechanism based on fuzzy Petri nets; and explanation of reasoning process through the use of hierarchical fuzzy Petri nets. FPNES offer several important benefits: (1) providing reasoning mechanism for uncertain and fuzzy information, (2) improving the efficiency of rule based reasoning by constructing relationship of concurrency among rule activation, and (3) explaining how to reach conclusions through the movements of tokens in fuzzy Petri nets. An application to the damage assessment of the Da-Shi bridge in Taiwan is used as an illustrative example of FPNES.

The paper presents a mechanism for normalizing fuzzy rules of a rule-based system, transforming the rules into a fuzzy Petri net, and answering queries based on the net with imprecise information. In addition, the authors present an algorithm utilizing the formalism of Petri nets to compute the degree of truth of the resulting answers to the queries. The degree of truth is computed in algebraic form based on the state equation which can be implemented in matrix computation in Petri nets.

Timing analysis is essential to the development of valid computer systems especially for a distributed real-time system. We present both static and dynamic analysis procedures to verify timing properties of distributed real-time systems using timing constraint Petri nets (TCPNs). With both derived static information and collected dynamic data, we can locate a faulty task in a distributed real-time system using a timing behavior cross reference task. After a faulty task has been pinpointed the causes of a timing constraint violation in a faulty task can be identified.

Based on system execution traces, this paper presents a dynamic approach for visualizing and debugging timing constraint violations occurring in distributed real-time systems. The system execution traces used for visualization and debugging are collected during the execution of a target program in such a way that its run-time behavior is not interfered with. This is made possible by our non-interference distributed real-time monitoring system which is capable of collecting system’s run-time traces by monitoring and fetching the data passing through the internal buses of a target system. After the run-time data has been collected, the visualization and debugging activities then proceeded. The timing behavior of a target program is visualized as two graphs—the Colored Process Interaction Graph (CPIG) and the Dedicated Colored Process Interaction Graph (DCPIG). The CPIG depicts the timing behavior of a target program by graphically representing interprocess relationships during their communication and synchronization. The DCPIG can reduce visualization and debugging complexity by focusing on the portion of a target program which has direct or indirect correspondence with an imposed timing constraint. With the help of the CPIG and the DCPIG, a timing analysis method is used for computing the system-related timing statistics and analyzing the causes of timing constraint violations. A visualization and debugging system, called VDS, has been implemented using OpenWindows on Sun-4’s/UNIX workstations.