Design of Front-End for Recommendation Systems: Towards a Hybrid Architecture

To provide personalized online shopping suggestions, recommendation systems play an increasingly important role in “closing a transaction”. Some leading online movie sales platforms, such as Netflix and Rotten Tomatoes, have exploited content-based recommendation approaches. However, the issue of insufficient information about features in item profiles may lead to less accurate recommendations. In this paper, we propose a recommendation method known as Collective Intelligence Social Tagging (CIST), which combines a content-based recommendation approach with a social tagging function based on crowd-sourcing. We used an online movie sales platform as a use-case of how a CIST approach could increase the accuracy of recommended results and the overall user experience. In order t0 understand the feasibility and satisfaction level for CIST, we conducted fifteen design interviews to first determine user-developer perspectives on CIST, and then collected their overall design input.

[1]  Daniel Roggen,et al.  Tagging human activities in video by crowdsourcing , 2013, ICMR.

[2]  Dan Conway,et al.  How to Recommend?: User Trust Factors in Movie Recommender Systems , 2017, IUI.

[3]  Kartik Hosanagar,et al.  Blockbuster Culture's Next Rise or Fall: The Impact of Recommender Systems on Sales Diversity , 2007, Manag. Sci..

[4]  Zehra Cataltepe,et al.  Content-Based Movie Recommendation Using Different Feature Sets , .

[5]  John Riedl,et al.  Tagsplanations: explaining recommendations using tags , 2009, IUI.

[6]  Din J. Wasem,et al.  Mining of Massive Datasets , 2014 .

[7]  Chun-Jung Chu,et al.  ASSOCIATION RULE MINING IN E-COMMERCE : A SURVEY Venkateswari S , 2011 .

[8]  Ravi S. Sharma,et al.  Evaluation measures for computer-based information systems , 1993 .

[9]  Vikas Sindhwani,et al.  Data Quality from Crowdsourcing: A Study of Annotation Selection Criteria , 2009, HLT-NAACL 2009.

[10]  Ahmed Mohammed K. Alsalama,et al.  A Hybrid Recommendation System Based On Association Rules , 2015 .

[11]  Samir Chatterjee,et al.  A Design Science Research Methodology for Information Systems Research , 2008 .

[12]  John Riedl,et al.  Recommender systems in e-commerce , 1999, EC '99.

[13]  Sonia Bergamaschi,et al.  Comparing LDA and LSA Topic Models for Content-Based Movie Recommendation Systems , 2014, WEBIST.

[14]  Louise Corti,et al.  Confidentiality and Informed Consent: Issues for Consideration in the Preservation of and Provision of Access to Qualitative Data Archives , 2000 .

[15]  Zehra Cataltepe,et al.  Feature selection for movie recommendation , 2016 .

[16]  G. Eysenbach,et al.  Ethical issues in qualitative research on internet communities , 2001, BMJ : British Medical Journal.

[17]  Ziad Salam Patrous,et al.  Evaluating Prediction Accuracy for Collaborative Filtering Algorithms in Recommender Systems , 2016 .

[18]  G. Giannachi,et al.  Digital humanities and crowdsourcing: an exploration , 2013 .

[19]  Ivo Blohm,et al.  How to Design an Internal Crowdsourcing System , 2017, ICIS.

[20]  Ravi S. Sharma,et al.  Case studies on the exploitation of crowd-sourcing with Web 2.0 functionalities , 2014, Int. J. Electron. Bus..

[21]  Vinjamuri Swathi,et al.  Music Recommendation System Using Association Rules , 2014 .

[22]  Jae Kyeong Kim,et al.  A literature review and classification of recommender systems research , 2012, Expert Syst. Appl..