On Segmentation of Interaction Values - Developing Concepts in Applied Intelligence - page 32

5Con lu on
We can evaluate the behaviors of interestingness measures by segmenting the
set of interestingness measures in the interaction graph. By using the capacity
function of Sugeno (corresponding to the parameter value of λ =0 . 5),atthe
first step, we have modeled the interaction between 40 interestingness measures
via the interaction matrix and the interaction graph. The observation of the
interaction between interestingness measures can be performed via the decrease
of interaction values (i.e., the threshold τ -interaction) or the segmentation of
interaction values. Based on the clusters of interstingness measures found, the
user can select the appropriated clusters of interestingness measures to evaluate
the quality of knwoledge represented in the form of association rules.
Acknowledgements
This publication was made possible through support provided by the IRD-DSF.
References
1. Agrawal, R., Srikant, R.: Fast algorithms for mining association rules in large
databases. In: Proceedings of 20th International Conference on Very Large Data
Bases, VLDB 1994, pp. 487-499 (1994)
2. Choquet, G.: Theory of capacities. Annales de l'Institut Fourier 5, 131-295 (1953)
3. Grabisch, M.: The application of fuzzy integrals in multicriteria decision making.
European Journal of Operational Research 89(3), 445-456 (1996)
4. Grabisch, M., Roubens, M.: An axiomatic approach to the concept of interaction
among players in cooperative games. International Journal of Game Theory (28),
547-565 (1999)
5. Huynh, X.-H., Guillet, F., Briand, H.: Evaluating interestingness measures with
linear correlation graph. In: Ali, M., Dapoigny, R. (eds.) IEA/AIE 2006. LNCS
(LNAI), vol. 4031, pp. 312-321. Springer, Heidelberg (2006)
6. Huynh, H.X.: Interestingness measures for association rules in a KDD process:
postprocessing of rules with ARQAT tool, PhD Thesis, Nantes University (2006)
7. Marichal, J.L.: Aggregation of interacting criteria by means of the discrete Cho-
quet integral, Aggregation Operators: New Trends and Applications. Studies in
Fuzziness and Soft Computing, vol. 97, pp. 224-244 (2002)
8. Murofushi, T., Soneda, S.: Techniques for reading fuzzy measures (iii): Interaction
index. In: Proceedings of the 9th Fuzzy Systems Symposium, pp. 693-696 (1993)
9. Sugeno, M.: Theory of fuzzy integrals and its applications, PhD Thesis, Tokyo
Institute of Technology (1974)
10. Takahagi, E.: On identification methods of λ -fuzzy measure identification using
weights and λ . Japanese Journal of Fuzzy Sets and Systems 12(5), 665-676 (2000)
11. Lam, N.C., Huynh, H.X., Guillet, F.: On interestingness measures interaction. In:
IEEE RIVF 2008: Proceedings of the 6th IEEE International Conference on Com-
puter Sciences: Research & Innovation - Vision for the Future, Ho-chi-minh Ville,
Vietnam, pp. 161-166 (2008)