Collaborate Filtering using social media knowledge

Kasper Aaquist Johansen

AbstractThe goal of the thesis is to use social media knowledge for collaborative filtering creating a recommendation system that can be used to recommend publications from Issuu’s publication inventory, to a user. To do so a collaborative filtering model has been created using a dataset provided by Issuu, and data from both Facebook and Twitter has been mined and used to do collaborative filtering and content based filtering. Using the data mined from Facebook, we have looked at features found amongst the users that could be used to create augmented collaborative filtering using the collaborative filtering model created on the Issuu dataset. With the Twitter data we have looked into how to extract trending topics that could be used as a content based filter to recommend publications from Issuu’s publication inventory.
We evaluate the model created and look at the results of the recommendations based on both the collaborative filtering and content based filtering. Finally we include a discussion of the results achieved, alternative methods to achieve the same result or a better result, and what future work that can be done, along with a conclusion on the goal of the thesis.
TypeMaster's thesis [Industrial collaboration]
Year2014
PublisherTechnical University of Denmark, Department of Applied Mathematics and Computer Science
AddressMatematiktorvet, Building 303B, DK-2800 Kgs. Lyngby, Denmark, compute@compute.dtu.dk
SeriesDTU Compute M.Sc.-2014
NoteDTU supervisor: Ole Winther, olwi@dtu.dk, DTU Compute. Thesis not publicly available.
Publication linkhttp://www.compute.dtu.dk/English.aspx
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing


Back  ::  IMM Publications