In this study, we propose quantifying students' learning depth in video programming courses by analyzing their video annotations using natural language processing techniques and Bloom's Digital Taxonomy (BDT). Utilizing a video-based Learning Management System that allows direct annotation of video frames, we collected 3,014 annotations from 42 students enrolled in a “Data Management” course. We optimized Japanese morphological analysis tools by integrating a custom IT terminology dictionary to enhance the recognition of domain-specific terms. A specialized corpus was constructed, and a domain-specific Word2vec model was trained to expand the BDT classification table through word embeddings. This expansion enabled the comprehensive classification of keywords extracted from student annotations into cognitive process categories. The results revealed a clear shift from lower-order to higher-order cognitive processes as the course progressed, indicating significant changes in students' learning depth influenced by instructional content and teaching strategies. The method effectively captured dynamic trajectories of cognitive development, and these findings provide educators with valuable insights to conduct nuanced assessments and enhance personalized learning experiences. This study provides a practical method for quantifying the depth of learning in online programming education, advancing the understanding of students' cognitive processes, and supporting the improvement of teaching practices through data-driven adjustments to instructional design.

This research full paper describes a study on Video Position-Based Notes (VPB-Notes). Video-based learning environments have become increasingly prevalent in programming education, offering flexibility and personalized pacing. The VPB-Notes which allow students to capture thoughts, questions, and observations at specific video timestamps, have emerged as a promising tool to promote active learning and engagement. However, the lack of standardization in VPB-Notes and the presence of mixed code and natural language poses significant challenges for learning analysis using traditional natural language processing techniques. This study aims to propose a novel analytical framework for understanding the cognitive processes and engagement levels reflected in VPB-Notes taken by students in asynchronous video-based programming courses. The research questions focus on uncovering the key characteristics and patterns of VPB-Notes and investigating the relationships between VPB-Notes features and learners' performance outcomes. Our study combines qualitative analysis of VPB-Notes from four programming courses with quantitative analysis using correlation and cluster analysis. The qualitative analysis reveals that VPB- Notes exhibit contextual relevance, conciseness, dynamic structure, and code integration with commentary. It also proposes an analytical framework for categorizing notes across four dimensions. The quantitative analysis suggests a hierarchy of note features, types, and initiative associated with academic performance. In contrast, cluster analysis highlights the complexity of note portfolios and the influence of personal learning styles. Despite limitations in the automation of note analysis and the need for further validation, this study contributes new insights to the growing body of knowledge on note-taking in online programming education, offering valuable guidance for instructional design and personalized learning support.

In a prior study we introduced the Programmed Visual Contents Comparison Method (PVCC) for assessment of programming abilities related with Panoramic Understanding of Programming (PUP). With this method, by comparing two or more output pictures produced by programming samples (a question), a student must decide which one of the programs producing them is more difficult to build with programming, or, if the difficulty is similar for all of them. This study reported also the results of a test to evaluate this method's validity performed with groups of students of diverse fields. Validity was verified by comparing the initial programming ability reported by programming professors of these groups with the test results; this confirmed that the PVCC method worked well to find programming abilities related with PUP. In this paper we propose an enhancement to the PVCC Method based on the preparation of New Questions where two or more samples displaying both input data and output pictures are shown. By adding input data to output pictures and focus strictly on the programming processes needed to obtain these pictures from the provided input data, we aim to broaden the range of discernible programming abilities related with PUP. The results of a test performed to verify the suitability of New Questions performed with professors of programming is also reported.

In recent years, professionals in different fields have become able to do programming by using simplified software tools, as a consequence of this they are becoming able to understand programming in a general or “panoramic” way. This understanding is not evaluated by current programming abilities testing methods such as written paper tests or practical programming. This paper proposes a Programmed Visual Contents Comparison Method to assess programming ability, and additionally, a testing system based on this method. With this method, by comparing 2 displayed images and interactive animations produced by programming samples (a question) a subject must decide which one of the programs is more difficult to build with programming than the other, or, if the difficulty is similar for both of them. The validity of the method is confirmed by comparing the ability reported by programming teachers with the results of an experiment performed with a testing system.

A number of people are killed or injured due to sudden disaster such as fire, terrorism or earthquake recently all over the world. The majority of the cause of death in these disasters is an escape delay. So the most effective method to reduce victims at the time of the disaster is to offer evacuation support information to the victim. To solve this problem, we are developing ERESS (Emergency Rescue Evacuation Support System) as a system that provides evacuation support in real time. In this paper, we propose a disaster detection and identification of disaster position using area information.

In this paper, we describe a test generation module to support personalized phoneme-based training in listening learning of English vocabulary. On the basis of the personalized phoneme error patterns detected by an error detection module, the test generation module would generate personalized listening tests. An evaluation experiment was conducted for evaluating the module.

Previously, the authors have proposed the Emergency Evacuation Support System (ERESS) for reducing disaster damage. The ERESS runs under Mobile Ad-hoc Networks (MANET) and primarily aims to reduce the number of victims in panic-type disasters. This system uses ERESS Mobile Terminals (EMT) which are mobile terminals assuming smartphones and tablets. EMT is provided with an advanced disaster detection algorithm and acquires sensor information such as acceleration, direction difference and walking steps. However, disaster detection of conventional ERESS is needed the information of all members existing on the same floor. In this paper, we propose a new disaster detection method using the positioning area information. Location information of the terminal holders are obtained by Radio Frequency Identification (RFID). In this method, we separate a number of areas in the floor and detect a disaster by the information of evacuees in the presence of each area. We show the effectiveness of the proposed method by panic-type experiments.

Nowadays, a lot of people die or are injured due to sudden disasters. In these disasters, it is important to discover them early and evacuate people quickly. To cope with these problems, we are developing ERESS (Emergency Rescue Evacuation Support System). Mobile terminals with this system exchange their information that associate with their sensors among them by MANET. When they detect a disaster occurring, they search an appropriate evacuation route in real time, and can reduce the number of victims. To realize this system, we need to analyze evacuees' behavior and establish a new judgment method of the disaster occurrence. Therefore, we are developing a system named Pre-ERESS, for collecting and analyzing the evacuees' behavior in emergency. We'll demonstrate this system.

Disastrous accidents (fires, earthquakes, chemical releases, terrorist attacks, indiscriminate gun-shootings, etc.) in local area (urban cities, residential buildings, department stores, etc.) usually result in tragic consequences for people and environments. Such accidents have clearly shown that the necessity for reliable systems supporting evacuation is quite urgent. Particularly choosing safety evacuation route is the most important for persons in affected area. In this paper we propose Emergency Urgent Communication -Evacuation Support (EUC-ES) system based on Mobile Ad-hoc NETworks (MANETs) among many mobile terminals. This system will works anywhere regardless of situations (indoor, urban, basement, etc.) as far as mobile terminals grasp their own positions. From the computer simulation results, it is shown that EUC-ES can evacuate affected persons in safety from the fire which occurred accidentally by using an algorithm to search for evacuation routes.

A novel packet switch architecture is proposed based on a channel grouped virtual circuit (CG-VC) scheme for high-speed communication networks. The case is considered in which there are several parallel channels between two switching nodes. The CG-VC scheme allows a packet to select an available channel in such a way as to avoid congested channels using a simple, hardware-based, self-routing mechanism. As a result, the architecture provides efficient channel use and low switching delay. The performance of the proposed switch is evaluated by simulation. The results show the proposed switch significantly improves channel utilization and switching delay compared with conventional single-channel allocation transmission policies.>