Please use this identifier to cite or link to this item: http://hdl.handle.net/11328/4052
Title: A 2020 perspective on “Scalable modelling and recommendation using wiki-based crowdsourced repositories:” Fairness, scalability, and real-time recommendation
Keywords: Algorithmic fairness
Crowdsourcing
Data stream mining
Profiling
Recommendation
Scalability
Issue Date: Apr-2020
Citation: Leal, F., Veloso, B., Malheiro, B., González-Vélez, H., & Burguillo, J. C. (2020). A 2020 perspective on “Scalable modelling and recommendation using wiki-based crowdsourced repositories:” Fairness, scalability, and real-time recommendation. Electronic Commerce Research and Applications, 40(March-April 2020), 100951. https://doi.org/10.1016/j.elerap.2020.100951. Repositório Institucional UPT. http://hdl.handle.net/11328/4052
Abstract: Wiki-based crowdsourced data sources generally lack reliability, as their provenance is not intrinsically marshalled. By using recommendation, one may arguably assess the reliability of wiki-based repositories in order to identify the most interesting articles for a given domain. In this commentary, we explore current trends in scalable modelling and recommendation methods based on side information such as the quality and popularity of wiki articles. The systematic parallelization of such profiling and recommendation algorithms allows the concurrent processing of distributed crowdsourced Wikidata repositories. These algorithms, which perform incremental updating, need further research to improve the performance and generate up-to-date high-quality recommendations. This article builds upon our previous work (Leal et al., 2019) by extending the literature review and identifying important trends and challenges pertaining to crowdsourcing platforms, particularly those of Wikidata provenance.
URI: http://hdl.handle.net/11328/4052
ISSN: 1567-4223 (Print)
Appears in Collections:REMIT – Artigos em Revistas Internacionais / Papers in International Journals

Files in This Item:
File Description SizeFormat 
A 2020 perspective on Scalable modelling and recommendation.pdf125.62 kBAdobe PDFView/Open    Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.