Evaluating adaptive and generative AI-based feedback and recommendations in a knowledge-graph-integrated programming learning system

Teaching programming skills is not only required for computer related departments but through the area of engineering and natural science. Moreover recently teaching programming skill is emphasized in software education for primary schools and secondary schools. Since programming ability is considered an essencial element of national competitiveness, we need programming learning system which alleviates the difficulty. We implemented Blockly webc Programming Convergent Learning System which is based on the graphic tools called Blockly by Google. Inside system problem sets for the programming beginners are embedded in the system. These problem sets are gone under more than 20 years verification and these problem sets may be used to help beginning programmers escape novice coder in short time. Blockly webc Programming Convergent Learning System together with already developed Simple Visual Language2 Programming Learning System is expected to play an important role as a programming learning system for the beginners. •

The purpose of this research is to develop a programming learning system using the completion strategy for beginners. The completion strategy was proved to have feasible result for beginners. Based on the previous works, most of programming tutors require students to write programs and then to evaluate and explain the program correctness. This “program generation” approach is not suitable to beginners. Therefore, the purpose of this research adopts the completion strategy to have beginners learn programming. The completion strategy is to utilize the well-designed programs to let students engage in completing, modifying, and extending their program. This research first investigates the programming learning theories; then initiates the system structure of the programming learning system. Template technique is used to implement the proposed system.

In recent years, Computer programming has become a critical subject, and it is a basic literacy in the digital age. However, learning programming skills is not an easy task as supported by many studies. On the other hand, many studies have concluded that student question generation has a positive effect on students learning, but few studies have geared toward supporting its use in programming classes. Therefore, this study aimed to develop a programming learning system named Peer-Interaction Programming Learning System based on a question generation strategy. The system was built by adding and intergrading learning assistance functions by system initiative into existing open-source QaA system. The usability of the system was examined for an undergraduate computer science course: Introduction to Computer Programming (n = 33). In this paper, we reported the design of the system and its user interface, identified several related systems, discussed our motivation and underlying teaching philosophy.

Along with the trend of computer technology, programming skill becomes the core competency of the future world, which highlights the importance of programming education. However, most of the related literature indicated that programming is difficult to be tackled for both learners who learn it and teachers who teach it. Despite of the importance, there are only a few cases in which teachers adopt system, which is made by programming, to teach students to design how to program. In this study, an online programming learning system was proposed to improve students' learning achievement, self-regulated learning ability and teachers' teaching experiences through several theory-based functions such as real-time screen-sharing, synchronous demonstration, and students' portfolio monitoring. To explore the learning efficiency of the proposed system, a one-group pretest-posttest design was performed in which 45 freshmen IT students were participants. The results have shown that meta-cognitive strategy and the programming achievement have improved. In the future, researchers will continue to adjust LETJS (Learning with Entertainment and Training in JavaScript) system and teaching strategy, taking an innovative approach to reach the goal of improving self-regulated action control strategy.

In recent years, Computer programming has become a very important subject, and it is a basic literacy in the digital age. But learning programming skills is not an easy task as supported by many studies. On the other hand, many studies have concluded that student question generation has a positive effect on students' learning. However, few studies have geared toward supporting its use in programming classes. Therefore, this Study aimed to develop a programming learning system named Peer-Interaction Programming Learning System based on a question generation strategy. The system was built by adding and intergrading learning assistance functions by system initiative into existing open-source Q&A system. In this paper, we reported the design of the system and its user interface, identified several related systems, discussed our motivation and underlying teaching philosophy.

In this paper, we propose a programming learning system which incorporates a concept of small steps to nurture the computational thinking. While the project-based programming environment allows children to reach an unlimited goal of their achievements, most of learners often suffer from finding a meaningful project to work and coming up with a way of building programs for their project. In order to scaffold their programming experience, our system gives a step-by-step procedure to build a complete meaningful project, where the entire project is decomposed into small steps in advance. In addition, the system requires learners to ask their peer to check if an ongoing step is correctly completed before proceeding to the next step. In this way, our system allows them to experience the repeated cycle of the computational thinking process while encouraging them to interact each other. We also show our empirical findings obtained by applying this system to a programming workshop. In this workshop, 11 small steps to build a game project were provided to participants with our iPad application. As a result, we observed that they could complete the given project regardless of their programming experience and the system gave them an opportunity to interact with others while they were doing programming.

This paper describes a new programming learning system. It consists of a Web-based flowchart application and Light Emitting Diode (LED) display kits. The LED kits produce various animations or static images according to a flow chart. The flowchart tool allows students to manipulate program elements through a Graphical User Interface (GUI). Survey results show that the proposed system makes learning how to program less challenging and more interesting for beginners. Encouraged by these initial findings, the authors are developing a system with add-on features including sensor accessory options and compatibility with well known LEGO® products.

Machine learning (ML) is implementing artificial intelligence (AI) research within medicine that has made dramatic progress in recent years. In addition to standard treatments, the role of complementary and alternative medicine should be mentioned. Traditional Thai medicine has received growing acceptance as a complementary approach to modern medicine by using local herbs. A vast amount of Thai herbal knowledge and information is freely available on the Internet. The reader must evaluate each website and decide to use trustworthy and appropriate information. This study aimed to acquire Thai herbal knowledge recorded in the Thai language system on the Internet by scraping websites using programming techniques. The knowledge was extracted with programming, and the types of Thai herbs were classified corresponding to target symptoms by the machine learning algorithm. The ML method organized the process when sufficient achievement was reached in order to give reliable and high accuracy results from the training data set. The validation of extracted knowledge was achieved by using the part-of-speech tag patterns analysis. This study showed that the programming and machine learning system was appropriate for obtaining and classifying Thai herbal medicines knowledge.

Cognitive ability is closely associated with the acquisition of programming skills, and enhancing learners’ cognitive ability is a crucial factor in improving the efficacy of programming education. Adaptive feedback strategies can provide learners with personalized support based on their learning context, which helps to stimulate their interest and improve learning outcomes. Nevertheless, it remains unclear whether adaptive feedback can enhance the cognitive ability of programming learners. This study applies adaptive feedback strategies to an introductory programming course by designing a quasi-experiment to analyze and reveal the effects of adaptive feedback on the cognitive ability to program learners from multiple modalities, including physiological, psychological, and behavioral data. Sixty-five first-year university students were randomly assigned to either experimental or control groups. The experimental group received adaptive feedback during the programming learning process and the control group received non-differential feedback. The findings demonstrated that adaptive feedback significantly enhances the cognitive ability of programming learners. The experimental group demonstrated superior performance in cognitive ability tests, programming examinations, and programming self-efficacy. Furthermore, adaptive feedback was found to markedly improve the cognitive processing of learners, as evidenced by the amplitude and latency of the P300 component in EEG signals, key-press reaction times and accuracy rates.

Prior work suggests that novice programmers are greatly impacted by the feedback provided by their programming environments. While some research has examined the impact of feedback on student learning in programming, there is no work (to our knowledge) that examines the impact of adaptive immediate feedback within programming environments on students' desire to persist in computer science (CS). In this paper, we integrate an adaptive immediate feedback (AIF) system into a block-based programming environment. Our AIF system is novel because it provides personalized positive and corrective feedback to students in real time as they work. In a controlled pilot study with novice high-school programmers, we show that our AIF system significantly increased students' intentions to persist in CS, and that students using AIF had greater engagement (as measured by their lower idle time) compared to students in the control condition. Further, we found evidence that the AIF system may improve student learning, as measured by student performance in a subsequent task without AIF. In interviews, students found the system fun and helpful, and reported feeling more focused and engaged. We hope this paper spurs more research on adaptive immediate feedback and the impact of programming environments on students' intentions to persist in CS.

Technology enhanced learning environments usually track a variety of data about students’ behaviour while students are learning. These data can be used to infer valuable information about how students learn as well as about their characteristics. This chapter focuses on the consideration of learning styles in technology enhanced learning. Considering students’ learning styles in technology enhanced learning can have many benefits for students such as providing them with personalized recommendations and advice based on their learning styles. In this chapter, we introduce an architecture that aims at dynamically identifying students’ learning styles from their behaviour in a learning system by frequently checking students’ behaviour and updating their learning styles based on their recent behaviour. Such a dynamic student modelling approach enables systems to incrementally learn students’ learning styles, identify and consider exceptional behaviour of students, and respond to changes in students’ learning styles by updating the student model respectively. The proposed architecture has been developed with only few dependencies to the learning system, making it possible to easily adjust and use the architecture for different learning systems. In this chapter, the integration of the proposed architecture into a particular learning system is demonstrated. Furthermore, an adaptivity module has been developed to show the benefits of the proposed architecture. This adaptivity module accesses the information about students’ learning styles to provide students with adaptive feedback about their learning styles as well as about how to improve their learning processes considering their learning styles and their courses.

This research proposes an evaluation model to measure a computer programming learning system success, for helping students to overcome learning difficulties. Teachers use programming learning environments, as computer-assisted instruction systems. The proposed model is supported in a the theory of information systems success evaluation, and in the characteristics of a virtual programming learning tool. The used tool was a web-based virtual robot learning system. Results presented here suggest the main features that this type of solution should include.

In education, computer learning systems make wide use of pedagogical agents to simulate a tutor and/or mimic tutoring interaction, as well as offering just-in-time and adaptive feedback. Although the theoretical aspect of the pedagogical agents has been well-documented in the literature, relatively fewer efforts have been made on how a pedagogical agent should be implemented in a real multimedia computerized simulation learning environment. In this paper, we propose a pedagogical agent architecture and implement it in the multimedia medical simulation web-based learning system call HINTS to further facilitate students' learning and thereby make the HINTS a more effective educational tool. A preliminary students' performance evaluation result is reported. Some experiments have been conducted and the results show that the pedagogical agent indeed help the students in their learning process.

In medical education, pedagogical agents are widely used by computer learning systems to simulate tutors and/or mimic tutoring interactions, as well as offering just-in-time and adaptive feedback. Although the theoretical aspect of the pedagogical agents has been well-documented in literature, relatively fewer efforts have been made on how a pedagogical agent should be implemented in a real multimedia computerized simulation learning environment. In this paper, we propose a general purpose pedagogical agent architecture and implement it in the multimedia medical simulation Web-based learning system called health information network teaching system (HINTS) to further facilitate students' learning and thereby make the HINTS a more helpful educational tool. Our focus is the design of the general purpose pedagogical architecture and its implementation in a multimedia computerized simulation learning environment. A preliminary students' performance evaluation result is also reported. We analyzed how to evaluate the students' performance and how the hints were given by the pedagogical agent. The system has been installed in the National Cheng Kung University Medical Center, Tainan, Taiwan for trial purposes. Some experiments have been conducted and the results have shown that the pedagogical agent indeed help the students in their learning process.

Providing adaptive feedback to learners engaging in collaborative learning activities is one research topic in the development of intelligent tutoring systems. However, there is a need to investigate how to systematically evaluate a learner’s activities and provide feedback on them. The present study investigates how emotional states, detected through facial recognition, can be utilized to capture the learning process in a simple jigsaw-type collaborative task. It was predicted that when learners argue with each other and reason deeply, they may experience several emotional states such as positive and negative states. The results show that when learners work harder on developing a mutual understanding through conflictive interaction, negative emotions can be used to predict this process. This study contributes to the knowledge of how emotional states detected by facial recognition technology can be applied to predict learning process in conflictive tasks. Moreover, these empirical results will impact the development of adaptive feedback mechanisms for intelligent tutoring systems for collaborative learning.