| dc.description.abstract | The Distance Learning environments (EAD) are becoming more present in the daily lives of teachers and students, either solely in the distance format (e-learning) or as a support tool to presencial teaching (b-learning). Despite the increase in its use, the use of these tools should not only represent a digital storage mean and data sharing. They should also be useful for both the student and the teacher in the learning mediation process. In this context, monitoring of activities in these tools is very important to make its use more efficient. Just collecting and summarizing the logs of the student's activities in a quantitative manner - e.g access number - to further condense them in reports, may not provide enough information to justify the use of these environments. It is also important to assess qualitative issues, by identifying characteristics that enable personalized learning. Another problem found is that for each of these existing environments, the activities logs are stored in their own format and structure of each software. This, in turn, makes it difficult to create standardized analysis tools. In the papers related described in Chapter 3 it is possible to check that, despite using the same theoretical base (socio-historical theory) to support educational interactions, all works analyze qualitatively only communication tools such as forum chats, messages, e-mail, bypassing the rest of the existing resources in AVAs. However, educational interactions also occur through other technologies; for example by providing a PDF file or page content to the student, the teacher will be interacting with this student, even asynchronously. Another example can be represented by an activity which when delivered by the student he may receive feedback from the teacher. Still regarding the online mediation process, another point that was considered is that in online environments, some students may not use the communication tools, or use it in low frequency, but at the same time, it presents good grades and access features made available by the teacher. In this case, monitoring of the mediation process by the teacher is compromised if he sees only records of communication tools. However, by seeing the interactions of students in the rest of the resources the teacher can have a more complete overview on the state of the student. Based on what was mentioned above, the present study will use as a theoretical basis the socio-historical approach and the Theory of Speech Acts. Once the interaction process is grounded based on the first theory, it will be presented a classification model of the illocutionary acts to the interactions of the users in EAD environments, in other words, it will be identified the intentionality of an action performed by the user. For this, the model will make use of Web Semantic technologies, educational standards of records and Bayesian network for classification of speech acts. At the end of the study, it will be presented the results of the experiments performed to such model. Also, for future work it is foreseen the use of this model to identify the category of mediation that the student is found; this will be done starting from the assumption that speech acts classified in this study is used as a basis for such proposal (which is explained in Chapter 4.1 "Comparative analysis between TAF and the TSH"). The text initially addresses the socio-historical theory, the Theory of Speech Acts, Virtual Learning Environments and Interaction of Standardized Records, as well as, the Web Semantic and Bayesian networks. In the next chapter are seen the work that makes use of socio-historical theory and its relation to the mediation process in virtual learning environments. In Chapter 4, it is presented the Pedagogical Interaction Model as well as information on the developed prototype. The paper ends with the presentation of the final considerations on the proposal for future work. | en |
