Personalized Fashion Advice - Smart Information Systems: Computational Intelligence for Real-Life Applications - page 231

Systems using Bagging replicate instances of objects to better account for unbalanced
classes. For instance, we may assume that a majority of customers prefers a certain
subset of articles. In contrast, a minority of customers may dislike these. Replicating
customers of theminority supports the systemnot to suggest the articles' subset which
the minority dislikes. On the other hand, Boosting simultaneously learns models
with varying parametrization. Thus, systems obtain more robustness by considering
different aspects. Finally, we seek to support the systems as new articles enter the
collections. Applying similar clustering techniques to the collection of articles allows
the system to detect similarity patterns among them. Having established both arti-
cle and item clusters, systems may learn models on their interactions. For instance,
systems may detect that users of user cluster A are particularly likely to buy articles
of item cluster B . Observing preferences of user clusters for particular item clusters
enables the system to better control the selection process.
Acknowledgments This workwas funded by the Federal Ministry of Economic Affairs and Energy
(BMWi) under funding reference number KF2086104KM3.
References
1. G. Adomavicius, B. Mobasher, F. Ricci, A. Tuzhilin, Context-aware recommender systems. AI
Mag. 32 (3), 67-80 (2011)
2. D.W. Aha, D. Kibler, M.K. Albert, Instance-based learning algorithms. Mach. Learn. 6 (1),
37-66 (1991)
3. J. Baumgartner, You Are What You Wear: What Your Clothes Reveal About You (Da Capo
Lifelong Books, Boston, 2012)
4. R.M. Bell, Y. Koren, Lessons from the Netflix prize challenge. SIGKDD Explor. Newsl. 9 (2),
75-79 (2007)
5. Y. Bengio. Deep learning of representations for unsupervised and transfer learning, in JMLR:
Workshop and Conference Proceedings , pp. 17-37 (2012)
6. J. Bennett, S. Lanning, N. Netflix, The Netflix prize, in KDD Cup and Workshop in Conjunction
with KDD (2007)
7. D. Bridge, M.H. Göker, L. McGinty, B. Smyth, Case-based recommender systems. Knowl.
Eng. Rev. 20 (9), 315-320 (2005)
8. P. Brusilovsky, E. Millan, User models for adaptive hypermedia and adaptive educational
systems, in The Adaptive Web: Methods and Strategies of Web Personalization , Chapter 1, ed.
by P. Brusilovsky, A. Kobsa, W. Nejdl (Springer, Berlin, 2007), pp. 3-53
9. R. Burke, Knowledge-based recommender systems, in Encyclopedia of Library and Informa-
tion Systems (Marcel Dekker, New York, 2000), p. 2000
10. F. Cacheda, V. Carneiro, D. Fernández, V. Formoso, Comparison of collaborative filtering
algorithms: limitations of current techniques and proposals for scalable, high-performance
recommender systems. ACM Trans. Web 5 (1), 2:1-2:33 (2011)
11. M.A. Casselman-Dickson, M.L. Damhorst, Female bicyclists and interest in dress. Cloth. Text.
Res. J. 4 , 7-17 (1993)
12. M.-H. Chae, C. Black, J. Heitmeyer, Pre-purchase and post-purchase satisfaction and fashion
involvement of female tennis wear consumers. Int. J. Consum. Stud. 30 , 25-33 (2005)
13. L. Chen, P. Pearl, Critiquing-based recommenders: survey and emerging trends. User Model.
User-Adapt. Interact. 22 (1-2), 125-150 (2012)
14. M.R. DeLong, B. Minshall, K. Larntz, Use of schema for evaluating consumer response to an
apparel product. Cloth. Text. Res. J. 5 , 17-26 (1986)