Information Technology Reference
In-Depth Information
•
In the algorithms introduced in this chapter, the diversity is modeled by the dis-
tance or dissimilarity between any two documents. However, this might be incon-
sistent with the real behavior of users. The reason is that users usually will not
check every document in the result list, instead, they browse the ranking results in
a top down manner. As a consequence, it would make more sense to define diver-
sity as the relationship between a given document and all the documents ranked
before it.
•
In all the algorithms introduced in this chapter, the relationship is pairwise: either
similarity, dissimilarity, or preference. Accordingly, a matrix (or graph) is used
to model the relationship. However, in real applications, there are some other re-
lationships that may go beyond “pairwise”. For example, all the webpages in the
same website have a multi-way relationship. It is more appropriate to use tensor
(or hypergraph) to model such multi-way relationships. Accordingly, the rela-
tional ranking framework should be upgraded (note that tensor and hypergraph
can contain matrix and graph as special cases).
References
1. Agrawal, R., Gollapudi, S., Halverson, A., Ieong, S.: Diversifying search results. In: Proceed-
ings of the 2nd ACM International Conference on Web Search and Data Mining (WSDM
2009), pp. 5-14 (2009)
2. Clarke, C.L., Kolla, M., Cormack, G.V., Vechtomova, O., Ashkan, A., Buttcher, S., MacKin-
non, I.: Novelty and diversity in information retrieval evaluation. In: Proceedings of the 31st
Annual International ACM SIGIR Conference on Research and Development in Information
Retrieval (SIGIR 2008), pp. 659-666 (2008)
3. Craswell, N., Hawking, D., Wilkinson, R., Wu, M.: Overview of the trec 2003 web track. In:
Proceedings of the 12th Text Retrieval Conference (TREC 2003), pp. 78-92 (2003)
4. Gollapudi, S., Sharma, A.: An axiomatic approach for result diversification. In: Proceedings
of the 18th International Conference on World Wide Web (WWW 2009), pp. 381-390 (2009)
5. Qin, T., Liu, T.Y., Zhang, X.D., Chen, Z., Ma, W.Y.: A study of relevance propagation for web
search. In: Proceedings of the 28th Annual International ACM SIGIR Conference on Research
and Development in Information Retrieval (SIGIR 2005), pp. 408-415 (2005)
6. Qin, T., Liu, T.Y., Zhang, X.D., Wang, D., Li, H.: Learning to rank relational objects and
its application to web search. In: Proceedings of the 17th International Conference on World
Wide Web (WWW 2008), pp. 407-416 (2008)
7. Qin, T., Liu, T.Y., Zhang, X.D., Wang, D.S., Li, H.: Global ranking using continuous condi-
tional random fields. In: Advances in Neural Information Processing Systems 21 (NIPS 2008),
pp. 1281-1288 (2009)
8. Radlinski, F., Kleinberg, R., Joachims, T.: Learning diverse rankings with multi-armed ban-
dits. In: Proceedings of the 25th International Conference on Machine Learning (ICML 2008),
pp. 784-791 (2008)
9. Yue, Y., Joachims, T.: Predicting diverse subsets using structural SVM. In: Proceedings of the
25th International Conference on Machine Learning (ICML 2008), pp. 1224-1231 (2008)
