Universidade Federal do Rio Grande do Sul (BRAZIL)
About this paper:
Appears in: ICERI2016 Proceedings
Publication year: 2016
Pages: 8654-8663
ISBN: 978-84-617-5895-1
ISSN: 2340-1095
doi: 10.21125/iceri.2016.0959
Conference name: 9th annual International Conference of Education, Research and Innovation
Dates: 14-16 November, 2016
Location: Seville, Spain
This study aims to present the RecETC, which is a Content Recommender created for a Collective Text Editor (CTE). The goal was to address an emerging need identified in Collective Digital Writing. Clearly the popularization of the Internet has enhanced the construction and availability of materials, this generates an unmanageable volume of information on various topics on a daily basis. Searching, selecting, and categorizing references has become a huge challenge both for students and teachers. For the latter, the challenge also extends to the task of recommending good and relevant supporting materials. This research emerged from educational practices developed in an online CTE which enables the construction of Collective Digital Writing (CDW) at different times by subjects in geographically disperse locations. It is often difficult to search for and select relevant materials from reliable reference sources to aid in collective writing. This is exacerbated when subjects are immersed in the digital network and exposed to an endless web of hyperlinks. In this scenario, users easily lose their research focus and often find themselves without research parameters, which limits their ability to continue. Therefore, a tool to mine and recommend different materials in various formats to authors participating in a collective production was developed to address this need, based on studies of Recommender Systems. This tool, which is called RecETC, is an integrated content recommender for CTEs. The recommendations are made based on content developed by the authors and, through the use of content-based filtering techniques (CBF) and Collaborative Filtering (CF), the system stores author’s records in order to present personalized recommendations. Because Collective Digital Writing is a dynamic process, each time the recommender is triggered it performs an updated search, considering all of the content that has been written. However, if the authors would like to search and select references before producing any content, they can register keywords for the recommender to search and recommend sources related to the requested topic. A teacher can also register terms if they would like to adapt or improve the recommender’s results. The tool, which is integrated and activated within the CTE, was validated in undergraduate courses, graduate, and extension courses at the Federal University of Rio Grande do Sul (UFRGS). According to the participants, the content recommender was useful for finding objective references from reputable sources and also considering different formats of materials, within the CTE itself. One of the major benefits was reducing the time needed to select materials. Moreover, it decresed the loss of focus that occurs when the search is done on general Internet search sites. At the same time, teachers also found advantages to using content recommender, because it can suggest/adapt materials according to the way the writing develops over time. Therefore, RecETC has improved the CDW process by creating opportunities for users to remain in the editor and present a set of references to qualify textual production.
Collective Digital Writing, Content Recommendation Technology, Collective Text Editor, educational practices.