Aggregating Interestingness Measures in Associative Classifiers

• Matheus Freitas da Silva UNESP
• Veronica Oliveira de Carvalho UNESP

DOI: https://doi.org/10.5753/eniac.2018.4405

Resumo

A classificação associativa, a qual vem sendo muito utilizada em diversos domínios, visa a obtenção de um modelo preditivo em que o processo é baseado na extração de regras de associação. A geração do modelo ocorre em etapas, sendo uma delas voltadas a ordenar e podar um conjunto de regras. No que se refere a ordenação, uma das soluções é ranquear as regras por meio de medidas objetivas (MOs). O critério de ordenação impacta a acurácia do classificador. Nos trabalhos da literatura as MOs são exploradas individualmente. Diante do exposto, este trabalho tem por objetivo explorar a agregação de medidas, em que várias MOs são consideradas ao mesmo tempo, no contexto de classificadores associativos.

Referências

[Abdelhamid et al. 2012] Abdelhamid, N., Ayesh, A., and Thabtah, F. (2012). An experimental study of three different rule ranking formulas in associative classification. In 2012 International Conference for Internet Technology and Secured Transactions, pages 795–800.

[Abdelhamid et al. 2016] Abdelhamid, N., Jabbar, A. A., and Thabtah, F. (2016). Associative classification common research challenges. In 45th International Conference on Parallel Processing Workshops, pages 432–437.

[Alwidian et al. 2018] Alwidian, J., Hammo, B. H., and Obeid, N. (2018). WCBA: Weighted classification based on association rules algorithm for breast cancer disease. Applied Soft Computing, 62:536–549.

[Azevedo and Jorge 2007] Azevedo, P. J. and Jorge, A. M. (2007). Comparing rule measures for predictive association rules. In Machine Learning: ECML 2007, pages 510–517.

[Bong 2014] Bong, K. K., J. M. Q. C. A. T. M. S. (2014). Selection and aggregation of interestingnes measures: A review. 59(1):146–166.

[Bouker et al. 2014] Bouker, S., Saidi, R., Yahia, S. B., and Nguifo, E. M. (2014). Mining undominated association rules through interestingness measures. International Journal on Artificial Intelligence Tools, 23(4):22p.

[Coenen 2004] Coenen, F. (2004). LUCS KDD implementation of CBA (Classification Based on Associations). [Online. Acesso em 05-06-2018].

[Fayyad and Irani 1993] Fayyad, U. M. and Irani, K. B. (1993). Multi-interval discretization of continuous-valued attributes for classification learning. In International Joint Conference on Artificial Intelligence, pages 1022–1029.

[Han et al. 2013] Han, J., Kamber, M., and Pei, J. (2013). Data Mining: Concepts and Techniques. 3 edition.

[Hernández-León et al. 2014] Hernández-León, R., Hernández-Palancar, J., Carrasco-Ochoa, J. A., and Martínez-Trinidad, J. F. (2014). Studying netconf in hybrid rule ordering strategies for associative classification. In Pattern Recognition, pages 51–60.

[Jalali-Heravi and Zaïane 2010] Jalali-Heravi, M. and Zaïane, O. R. (2010). A study on interestingness measures for associative classifiers. In Proceedings of the 2010 ACM Symposium on Applied Computing, pages 1039–1046.

[Kannan 2010] Kannan, S. (2010). An Integration of Association Rules and Classification: An Empirical Analysis. PhD thesis, Madurai Kamaraj University.

[Liu et al. 1998] Liu, B., Hsu, W., and Ma, Y. (1998). Integrating classification and association rule mining. In Proceedings of the 4th International Conference on Knowledge Discovery and Data Mining, pages 80–86.

[Ma et al. 2014] Ma, B., Zhang, H., Chen, G., Zhao, Y., and Baesens, B. (2014). Investigating associative classification for software fault prediction: An experimental perspective. International Journal of Software Engineering and Knowledge Engineering, 24(1):61–90.

[Moreno et al. 2016] Moreno, M. N., Segrera, S., López, V. F., Muñoz, M. D., and Sánchez, A. L. (2016). Web mining based framework for solving usual problems in recommender systems. A case study for movies’ recommendation. Neurocomputing, 176:72–80.

[Nandhini et al. 2015] Nandhini, M., Sivanandam, S. N., Rajalakshmi, M., and Sidheswaran, D. (2015). Enhancing the spam email classification accuracy using post processing techniques. 10(15):35125–35130.

[Nguyen Le et al. 2009] Nguyen Le, T. T., Huynh, H. X., and Guillet, F. (2009). Knowledge acquisition: Approaches, algorithms and applications. chapter Finding the Most Interesting Association Rules by Aggregating Objective Interestingness Measures, pages 40–49.

[Shao et al. 2017] Shao, Y., Liu, B., Li, G., and Wang, S. (2017). Software defect prediction based on class-association rules. In 2nd International Conference on Reliability Systems Engineering, pages 1–5.

[Singh et al. 2016] Singh, J., Kamra, A., and Singh, H. (2016). Prediction of heart diseases using associative classification. In 5th International Conference on Wireless Networks and Embedded Systems, pages 1–7.

[Tew et al. 2014] Tew, C., Giraud-Carrier, C., Tanner, K., and Burton, S. (2014). Behaviorbased clustering and analysis of interestingness measures for association rule mining. Data Mining and Knowledge Discovery, 28(4):1004–1045.

[Thabtah 2007] Thabtah, F. (2007). A review of associative classification mining. Knowledge Engineering Review, 22(1):37–65.

[Yang and Cui 2015] Yang, G. and Cui, X. (2015). A study of interestingness measures for associative classification on imbalanced data. In Trends and Applications in Knowledge Discovery and Data Mining, pages 141–151.

[Yang et al. 2009] Yang, G., Shimada, K., Mabu, S., and Hirasawa, K. (2009). A nonlinear model to rank association rules based on semantic similarity and genetic network programing. IEEJ Transactions on Electrical and Electronic Engineering, 4(2):248–256.

[Yin et al. 2018] Yin, C., Guo, Y., Yang, J., and Ren, X. (2018). A new recommendation system on the basis of consumer initiative decision based on an associative classification approach. Industrial Management and Data Systems, 118(1):188–203.