Smart Objects and Technologies for Social Good. Third International Conference, GOODTECHS 2017, Pisa, Italy, November 29-30, 2017, Proceedings

Research Article

Context-Aware Recommendations for Sustainable Wardrobes

  • @INPROCEEDINGS{10.1007/978-3-319-76111-4_6,
        author={Anders Kolstad and \O{}zlem \O{}zg\o{}bek and Jon Gulla and Simon Litlehamar},
        title={Context-Aware Recommendations for Sustainable Wardrobes},
        proceedings={Smart Objects and Technologies for Social Good. Third International Conference, GOODTECHS 2017, Pisa, Italy, November 29-30, 2017, Proceedings},
        proceedings_a={GOODTECHS},
        year={2018},
        month={3},
        keywords={Internet of Things Recommender systems Content-based recommendation Textile recycling Linked open data Bag of concepts},
        doi={10.1007/978-3-319-76111-4_6}
    }
    
  • Anders Kolstad
    Özlem Özgöbek
    Jon Gulla
    Simon Litlehamar
    Year: 2018
    Context-Aware Recommendations for Sustainable Wardrobes
    GOODTECHS
    Springer
    DOI: 10.1007/978-3-319-76111-4_6
Anders Kolstad,*, Özlem Özgöbek1,*, Jon Gulla1,*, Simon Litlehamar2,*
  • 1: Norwegian University of Science and Technology
  • 2: Accenture AS
*Contact email: anders.kolstad@accenture.com, ozlem.ozgobek@ntnu.no, jon.atle.gulla@ntnu.no, simon.litlehamar@accenture.com

Abstract

Through recycling textile waste, greenhouse gas emissions can drastically be reduced. Such textile recycling has become a lot easier with clothing retailers now starting to offer recycling checkpoints. Moreover, people today are often challenged by overloaded wardrobes and store many clothing items that they never use. In this paper, we describe an Internet of Things system that creates incentives for the users to recycle their clothes, benefiting the environmental sustainability. We propose a content-based recommendation approach that utilizes semantic web technologies and that leverages a set of context signals obtained from the system’s architecture, to recommend clothing items that might be relevant for the user to recycle. Experiments on a real-world dataset show that our proposed approach outperforms a baseline which does not utilize semantic web technologies.