Rough Sets in Incomplete Information Systems with Order Relations Under Lipski’s Approach
Rough sets and rule induction based on them are described in incomplete information tables where attribute values are ordered. We apply possible world semantics to an incomplete information table, as Lipski did in incomplete databases. The set of possible tables on a set of attributes is derived from the original incomplete information table. Rough sets, a pair of lower and upper approximations, are obtained from every possible table. An object is certainly included in an approximation when it is in the approximation in all possible tables, while an object is possibly included in an approximation when it is in the approximation in some possible tables. From this, we obtain certain and possible approximations. The actual approximation is greater than the certain one and less than the possible one. Finally, we obtain the approximation in the form of interval sets. There exists a gap between rough sets and rule induction from them. To bridge rough sets and rule induction, we give expressions that correspond to certain and possible approximations. The expressions consist of a pair of an object and a rule that the object supports. Consequently, four types of rule supports: certain and consistent, certain and inconsistent, possible and consistent, and possible and inconsistent supports, are obtained from the expressions. The formulae can be applied to the case where not only attributes used to approximate but also attributes approximated have a value with incomplete information. The results give a correctness criterion of rough sets and rule induction based on them in incomplete ordered information systems, as the results of Lipski’s work are so in incomplete databases.
KeywordsRough sets Rule induction Incomplete information systems Ordered domains Possible world semantics
The authors wish to thank the anonymous reviewers for their valuable comments.
- 3.Chen, Z., Shi, P., Liu, P., Pei, Z.: Criteria reduction of set-valued ordered decision system based on approximation quality. Int. J. Innov. Comput. Inf. Control 9(6), 2393–2404 (2013)Google Scholar
- 6.Greco, S., Matarazzo, B., Słowinski, R.: Handling missing values in rough set analysis of multi-attribute and multi-criteria decision problems. In: Zhong, N., Skowron, A., Ohsuga, S. (eds.) RSFDGrC 1999. LNCS, vol. 1711, pp. 146–157. Springer, Heidelberg (1999). doi: 10.1007/978-3-540-48061-7_19 CrossRefGoogle Scholar
- 8.Hu, M., Yao, Y.: Definability in incomplete information tables. In: Flores, V., Gomide, F., Janusz, A., Meneses, C., Miao, D., Peters, G., Ślęzak, D., Wang, G., Weber, R., Yao, Y. (eds.) IJCRS 2016. LNCS, vol. 9920, pp. 177–186. Springer, Cham (2016). doi: 10.1007/978-3-319-47160-0_16 CrossRefGoogle Scholar
- 9.Imielinski, T.: Incomplete information in logical databases. Data Eng. 12, 93–104 (1989)Google Scholar
- 16.Nakata, M., Sakai, H.: Applying rough sets to information tables containing missing values. In: Proceedings of 39th International Symposium on Multiple-Valued Logic, pp. 286–291. IEEE Computer Society Press (2009)Google Scholar
- 20.Parsons, S.: Addendum to current approaches to handling imperfect information in data and knowledge bases. IEEE Trans. Knowl. Data Eng. 10, 862 (1998)Google Scholar
- 22.Qi, Y., Sun, H., Yang, X., Song, Y., Sun, Q.: Approches to approximate distribution reduct in incomplete ordered decision system. J. Inf. Comput. Sci. 3(3), 189–198 (2008)Google Scholar
- 24.Sakai, H., Okuma, A.: Basic algorithms and tools for rough non-deterministic information analysis. In: Peters, J.F., Skowron, A., Grzymała-Busse, J.W., Kostek, B., Świniarski, R.W., Szczuka, M.S. (eds.) Transactions on Rough Sets I. LNCS, vol. 3100, pp. 209–231. Springer, Heidelberg (2004). doi: 10.1007/978-3-540-27794-1_10 CrossRefGoogle Scholar
- 25.Sakai, H., Ishibashi, R., Koba, K., Nakata, M.: Rules and apriori algorithm in non-deterministic information systems. In: Peters, J.F., Skowron, A., Rybiński, H. (eds.) Transactions on Rough Sets IX. LNCS, vol. 5390, pp. 328–350. Springer, Heidelberg (2008). doi: 10.1007/978-3-540-89876-4_18 CrossRefGoogle Scholar
- 30.Wang, H., Guan, Y., Huang, J., Shen, J.: Decision rules acquisition for inconsistent disjunctive set-valued ordered decision information systems. Math. Prob. Eng. 2015, Article ID 936340, 8 p. (2015)Google Scholar