A Model for Semi-Automatic Composition of Educational Content from Open Repositories of Learning Objects

Learning objects (LOs) repositories are important in building educational content and should allow search, retrieval and composition processes to be successfully developed to reach educational goals. However, such processes require so much time-consuming and not always provide the desired results. T...

Cijeli opis

Spremljeno u:
Bibliografski detalji
Glavni autori: Rodríguez Marín, Paula Andrea, Moreno Cadavid, Julián, Duque Méndez, Néstor Darío, Ovalle Carranza, Demetrio Arturo, Silveira, Ricardo
Format: Online
Jezik:spa
Izdano: REDIE es una publicación del Instituto de Investigación y Desarrollo Educativo (IIDE). 2014
Online pristup:https://redie.uabc.mx/index.php/redie/article/view/728
Oznake: Dodaj oznaku
Bez oznaka, Budi prvi tko označuje ovaj zapis!
Opis
Sažetak:Learning objects (LOs) repositories are important in building educational content and should allow search, retrieval and composition processes to be successfully developed to reach educational goals. However, such processes require so much time-consuming and not always provide the desired results. Thus, the aim of this paper is to propose a model for the semiautomatic composition of LOs, which are automatically recovered from open repositories. For the development of model, various text similarity measures are discussed, while for calibration and validation some comparison experiments were performed using the results obtained by teachers. Experimental results show that when using a value of k (number of LOs selected) of at least 3, the percentage of similarities between the model and such made by experts exceeds 75%. To conclude, it can be established that the model proposed allows teachers to save time and effort for LOs selection by performing a pre-filter process.