Incorporating Content and Context in Recommender Systems

• Author / Creator

  Stepan, Torin KS

• Recommender systems are a growing area of research that find practical applications in a variety of domains. Integrated library systems and location-based social networks can apply recommendation algorithms to assist their users in finding an item or location that suits their needs. With an ever-increasing variety of options to choose from, deciding on which book to read or movie to watch can become overwhelming. Recommender systems aid their users in the decision making process by providing a list of items likely to be relevant to the user's needs and interests. A persistent issue faced by recommender systems is a lack of data concerning the preferences of its users, known as the "cold-start" problem, which leads to poor recommendation quality, particularly for new users and items. To improve recommendation quality in the face of incomplete data, we propose several novel approaches for incorporating all available data into collaborative filtering algorithms.

• Subjects / Keywords

  • Fuzzy
  • Rural libraries
  • Social
  • Cold-start
  • Location based social networks
  • Fuzzy taste vector
  • Recommendation
  • Locations
  • Similarity
  • University digital libraries
  • K-nearest neighbors
  • Content
  • Recommender
  • Classifier
  • Hybrid
  • Collaborative filtering
  • Books
  • Movies
  • Context
  • Spatial
  • Temporal

• Graduation date

  Spring 2015

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/R3ZM2H

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.