LinkedVis: exploring social and semantic career recommendations

This paper presents LinkedVis, an interactive visual recommender system that combines social and semantic knowledge to produce career recommendations based on the LinkedIn API. A collaborative (social) approach is employed to identify professionals with similar career paths and produce personalized recommendations of both companies and roles. To unify semantically identical but lexically distinct entities and arrive at better user models, we employ lightweight natural language processing and entity resolution using semantic information from a variety of end-points on the web. Elements from the underlying recommendation algorithm are exposed through an interactive interface that allows users to manipulate different aspects of the algorithm and the data it operates on, allowing users to explore a variety of "what-if" scenarios around their current profile. We evaluate LinkedVis through leave-one-out accuracy and diversity experiments on a data corpus collected from 47 users and their LinkedIn connections, as well as through a supervised study of 27 users exploring their own profile and recommendations interactively. Results show that our approach outperforms a benchmark recommendation algorithm without semantic resolution in terms of accuracy and diversity, and that the ability to tweak recommendations interactively by adjusting profile item and social connection weights further improves predictive accuracy. Questionnaires on the user experience with the explanatory and interactive aspects of the application reveal very high user acceptance and satisfaction.

[1]  John Riedl,et al.  GroupLens: an open architecture for collaborative filtering of netnews , 1994, CSCW '94.

[2]  Tobias Höllerer,et al.  Modeling topic specific credibility on twitter , 2012, IUI '12.

[3]  Franca Garzotto,et al.  Comparative evaluation of recommender system quality , 2011, CHI Extended Abstracts.

[4]  Barry Smyth,et al.  Trust in recommender systems , 2005, IUI.

[5]  Tobias Höllerer,et al.  WiGipedia: A Tool for Improving Structured Data in Wikipedia , 2011, 2011 IEEE Fifth International Conference on Semantic Computing.

[6]  Lior Rokach,et al.  Recommender Systems Handbook , 2010 .

[7]  Boon Chong Lim,et al.  Word-of-mouth: The use of source expertise in the evaluation of familiar and unfamiliar brands , 2014 .

[8]  W. Bruce Croft,et al.  Discovering key concepts in verbose queries , 2008, SIGIR '08.

[9]  George Karypis,et al.  Evaluation of Item-Based Top-N Recommendation Algorithms , 2001, CIKM '01.

[10]  Raymond J. Mooney,et al.  Content-boosted collaborative filtering for improved recommendations , 2002, AAAI/IAAI.

[11]  Sibel Adali,et al.  Credibility in Context: An Analysis of Feature Distributions in Twitter , 2012, 2012 International Conference on Privacy, Security, Risk and Trust and 2012 International Confernece on Social Computing.

[12]  Robin Burke,et al.  Identifying Attack Models for Secure Recommendation , 2004 .

[13]  Anette Hulth,et al.  Improved Automatic Keyword Extraction Given More Linguistic Knowledge , 2003, EMNLP.

[14]  Jonathan L. Herlocker,et al.  Evaluating collaborative filtering recommender systems , 2004, TOIS.

[15]  Guy Shani,et al.  Evaluating Recommendation Systems , 2011, Recommender Systems Handbook.

[16]  Pattie Maes,et al.  Social information filtering: algorithms for automating “word of mouth” , 1995, CHI '95.

[17]  Judith Masthoff,et al.  Designing and Evaluating Explanations for Recommender Systems , 2011, Recommender Systems Handbook.

[18]  Kenneth Y. Goldberg,et al.  Eigentaste: A Constant Time Collaborative Filtering Algorithm , 2001, Information Retrieval.

[19]  Judith Masthoff,et al.  Effective explanations of recommendations: user-centered design , 2007, RecSys '07.

[20]  Michael S. Bernstein,et al.  Eddi: interactive topic-based browsing of social status streams , 2010, UIST.

[21]  Tobias Höllerer,et al.  TasteWeights: a visual interactive hybrid recommender system , 2012, RecSys.

[22]  Derek Lackaff,et al.  An Analysis of Topical Coverage of Wikipedia , 2008, J. Comput. Mediat. Commun..

[23]  Alfred Kobsa,et al.  Inspectability and control in social recommenders , 2012, RecSys.

[24]  John Riedl,et al.  Item-based collaborative filtering recommendation algorithms , 2001, WWW '01.

[25]  Robin D. Burke,et al.  Hybrid Recommender Systems: Survey and Experiments , 2002, User Modeling and User-Adapted Interaction.

[26]  Derry O'Sullivan,et al.  Improving Case-Based Recommendation: A Collaborative Filtering Approach , 2002, ECCBR.

[27]  David Heckerman,et al.  Empirical Analysis of Predictive Algorithms for Collaborative Filtering , 1998, UAI.