Achieving better recommendations with overclassification: Practical considerations

Onofrei George, Archip Alexandru

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

With the growth of the Internet, recommendations have become a fact of life and there is a big competition between companies to get as much market share as possible. This is achievable not only through fast to market solutions but also by constantly providing high quality services. Considering the speed of current technological advancements, it is impossible to conceive a world without some form of prediction. Traditionally, recommendations have been computed using the collaborative filtering method, but the growth of available information has pushed research into other directions such as content based, demographic based, utility-based, knowledge-based and hybrids. This paper provides an introduction to collaborative filtering and content based methods, describing the underlying concepts, and discusses possible improvements using overclassification. Individual results are presented and then compared with the overclassification approach.

Original languageEnglish
Title of host publication2016 20th International Conference on System Theory, Control and Computing, ICSTCC 2016 - Joint Conference of SINTES 20, SACCS 16, SIMSIS 20 - Proceedings
EditorsEmil Petre, Marius Brezovan
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages410-416
Number of pages7
ISBN (Electronic)9781509027200
DOIs
Publication statusPublished - 16 Dec 2016
Externally publishedYes
Event20th International Conference on System Theory, Control and Computing, ICSTCC 2016 - Sinaia, Romania
Duration: 13 Oct 201615 Oct 2016

Publication series

Name2016 20th International Conference on System Theory, Control and Computing, ICSTCC 2016 - Joint Conference of SINTES 20, SACCS 16, SIMSIS 20 - Proceedings

Conference

Conference20th International Conference on System Theory, Control and Computing, ICSTCC 2016
Country/TerritoryRomania
CitySinaia
Period13/10/1615/10/16

Keywords

  • Apriori Rocchio
  • Naive Bayes
  • collaborative filtering
  • content based
  • k nearest neighbor
  • overclassification
  • recommendation system

Fingerprint

Dive into the research topics of 'Achieving better recommendations with overclassification: Practical considerations'. Together they form a unique fingerprint.

Cite this