COLLABORATIVE PROJECT-BASED LEARNING WITH CONTENT RECOMMENDATION SUPPORT
This paper aims to investigate the possible benefits of using an active learning approach, Project-Based Learning (PBL), as a way to contribute to collaborative practices in a virtual learning environment. PBL is a student-centered learning method that emphasizes activities for the development of projects and is seen as an active learning approach in which knowledge is built through the development of experimental and analytical activities. During this process, students can make their own decisions and act alone or in groups. For the application of the proposed method, an educational activity was structured involving the development of a project from students' questionings in relation to a theme proposed by the teacher. The development of this project starts and ends in the classroom, however the intermediate stages can occur in other locations. Toth is a tool that was developed to support these activities, consisting in a responsive web tool for fostering collaboration between students during the development of their projects. The tool allows message exchange between participants and also the possibility of students to collaborate on projects from each other through predefined types of comments. When students are working on their projects, the tool suggests additional materials related to the subject as a way to assist students in their research processes. When teachers propose a general topic of research, they upload a “triggering text” to prompt students to the subject. This text is mined together with the students' writings, aiming at the extraction of relevant terms and keywords. The mining method is based on the n-simple distance graph model, in which nodes represent the main terms found in the text, and the edges used to link nodes represent adjacency information. The items returned from the web search (articles, books, images, videos) according to the terms coming from the mining tool are then suggested to students, making them aware of the availability of different material that may be interesting in their investigation. In addition to recommending items from the web, the recommendation system monitors the materials accessed and used by students, with the goal of considering this information in new recommendations. For the evaluation of the proposed method a quali-quantitative research was structured , with data collected from questionnaires, interviews with teachers and system logs. Preliminary results have shown that the ability to scan students' writings in the search for related contents has the potential to provide them with relevant recommendations. It has also been observed that the system capacity to prompt students to collaborate with their peers has steered up interaction and participation in the activities proposed.