In this paper, we propose a framework of Kit-Build Concept Map (we call it as KB map) where a concept map made by a learner can be diagnosed automatically. In this research, we have divided the task to make a concept map into two sub-tasks: (1) “segmentation task” where parts of the concept map (nodes and links) are extracted from learning resources and (2) “structuring task” where the extracted parts are integrated into a map. In the framework of kit-build concept map, a learner is given a set of parts of a concept map and then re-builds the concept map by combining the given parts. In this process, the segmentation task becomes a task of recognition of the given parts and the structuring task remains as it is. The concept map should be prepared beforehand by a teacher or domain expert. We call this map “goal map.” The necessary and sufficient parts (kit) are generated by decomposing the goal map. The parts are provided to learners, and then the learners are required to build concept maps (learner maps) by connecting the parts. Since the same parts are used both in the goal map and a learner map, it is possible to find defects in the learner map as the differences from the goal map. By overlaying several learners’ maps, then, a group map can be generated. By comparing the group map with the goal map, differences between the goal map and the group of the learners are detected. We have also realized procedure to re-examine the goal map based on the differences between the group map and the goal map. We have already developed a system that realized the framework of the KB map. Through a preliminary use of the system, we have confirmed that the system works along with the framework of KB map. Evaluation of learning effect is our future work.

Text similarity measurement compares text with available references to indicate the degree of similarity between those objects. There have been many studies of text similarity and resulting in various approaches and algorithms. This paper investigates four majors text similarity measurements which include String-based, Corpus-based, Knowledge-based, and Hybrid similarities. The results of the investigation showed that the semantic similarity approach is more rational in finding substantial relationship between texts.

This study investigates whether learners consider constraints while posing arithmetic word problems. Through log data from an interactive learning environment, we analyzed actions of 39 first grade elementary school students and conducted correlation analysis between the frequency of actions and validity of actions. The results show that the learners consider constraints while posing arithmetic word problems.

The advancement of computer and communication technologies has enabled researchers to conduct and analyze the learning process of posing problems. This study investigates what learners think while posing problems as sentence integration in terms of intermediate products as well as the posed problems as the resultant product. Problem posing as sentence integration defines the arithmetic word problem structure, and posing a problem is a task to satisfy all the constraints and requirements to build a valid structure. A previous study shows that, in problem posing as sentence integration for arithmetic word problems, learners try to satisfy a relatively large number of constraints in the posed problems. In contrast, this study focuses on the violation of constraints in the intermediate products while posing problems. The result shows that learners were inclined to avoid as many violated constraints as possible throughout the problem-posing process. Although learners tend to avoid the violated constraints, naturally, they cannot avoid some mistakes. Further analysis shows that learners actually have difficulty in fulfilling particular constraints while posing the problems. Based on this analysis, it is possible to detect the difficulty of learners’ actions from the model perspective. Hence, it is possible to give accurate feedback and appropriately support the learners.

In this paper, we describe the effects of using Kit-Build concept mapping (KB-mapping) method as a technology-enhanced support for the Reading Comprehension (RC) in English as Foreign Language (EFL) contexts. RC is a process that helps learners to become a more effective and efficient reader. It is an intentional, active and interactive activity that language learners experience in their daily working activities. RC of EFL is a significant research area in technology-enhanced learning. In order to clarify the effect of KB-mapping method, we compared the results with that of selective underlining (SU) strategy through the Comprehension Test (CT) and the Delayed Comprehension Test (DCT) that performed two weeks later. As the results, it is clarified that there is a noticed difference in the DCT scores, while there is no significant difference in the CT scores. It indicates that the use of KB-mapping method helps learners to retain their information for longer period of time. By doing more statistical analysis for the results of the Kit- Build Conditions (KB-conditions) group and comparing them with the map scores, we found that the learners could answer 76% of the questions whose answers were included in their learner’s maps. It was found that learners could recall 86% of the questions and that their answers were included in their learner’s maps. It indicates that the use of KB-mapping method helps learners to retain and recall more information compared with the SU strategy even after two weeks elapsed. In a follow-up questionnaire after the end of all experiments, it was revealed that participants thought that using KB-mapping was similar to SU for the CT just after the use, but KB-mapping was more useful in remembering information after a while, and it was more difficult to carry out. Participants liked to use it in RC tasks, but asked for more time to do it.