| Petrovic S., Kendall G., Yang Y.: A Tabu Search Approach for Graph-Structured Case Retrieval. In: Proceedings of the STarting Artificial Intelligence Researchers Symposium, |
.... as a weighted combination of each attribute value pair similarity, and it can be eciently computed [5] However, more structured representations can be necessary for solving more complex problems, e.g. feature terms [14] hierarchical decompositions [15] directed or non directed labeled graphs [7, 13]. In this paper, we focus on directed labeled graphs, i.e. directed graphs whose vertices and edges are associated with one or more labels, and consider the problem of characterizing and computing their similarity. Such labeled graphs provide a rich mean for modeling structured objects. In ....
....transformations are necessary to make the graphs isomorphic. However, all the above propositions miss some interesting features of our model, e.g. approximate edge matching (most focus on vertices) handling splits, exible tuning. It is also worth mentioning the recent work of Petrovic et al. [13] about graph retrieval in the eld of CBR. Their approach is quite similar to ours it expresses similarity of a mapping between vertices, and looks for the maximizing mapping but it is lacking some exibility with regard to our requirements: splits are not allowed (mappings are univalent ....
Sanja Petrovic, Graham Kendall, and Yong Yang. A Tabu Search Approach for Graph-Structured Case Retrieval. In proceedings of STAIRS 02, volume 78 of Frontiers in Arti cial Intelligence and Applications, pages 55-64. IOS Press, 2002.
No context found.
Petrovic S., Kendall G., Yang Y.: A Tabu Search Approach for Graph-Structured Case Retrieval. In: Proceedings of the STarting Artificial Intelligence Researchers Symposium,
No context found.
Petrovic S., Kendall G., Yang Y.: A Tabu Search Approach for Graph-Structured Case Retrieval. In: Proceedings of the STarting Artificial Intelligence Researchers Symposium,
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC