This is an open access article published by the IET, Chinese Association for Artificial Intelligence and Chongqing University of Technology under the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/)
Information tables having continuous domains are handled by neighborhood rough sets. Two approximations in complete information tables are extended to handle incomplete information. Consequently, four approximations are obtained: certain and possible lower ones and certain and possible upper ones without computational complexity. These extended approximations create the same results as the ones from possible world semantics by using possible indiscernibility relations. Therefore, the extension is justified. In complete information tables two types of single rules that an object supports are obtained: consistent and inconsistent ones. The single rule has low applicability. To increase applicability, a series of single rules are brought into one combined rule with an interval value. In incomplete information tables four kinds of single rules are obtained. From them, four kinds of combined rules are obtained: certain and consistent, possible and consistent, certain and inconsistent, and possible inconsistent ones. A combined rule has higher applicability than the single rules from which it is assembled.
References
-
-
1)
-
[23]. Zimányi, E., Pirotte, A.: ‘Imperfect information in relational databases’, in Motro, A., Smets, P. (Eds.): ‘Uncertainty management in information systems: from needs to solutions’ (Kluwer Academic Publishers, Dordrecht, The Netherlands, 1997), pp. 35–87.
-
2)
-
[11]. Nakata, M., Sakai, H.: ‘Applying rough sets to information tables containing missing values’. Proc. 39th Int. Symp. on Multiple-Valued Logic, Naha, Okinawa, Japan, 2009, Changchun, Jilin, China, 2009, pp. 286–291.
-
3)
-
[13]. Nakata, M., Sakai, H.: ‘Rough sets handling missing values probabilistically interpreted’, Lect. Notes Artif. Intell., 2005, 3641, pp. 325–334.
-
4)
-
[18]. Abiteboul, S., Hull, R., Vianu, V.: ‘Foundations of databases’ (Addison-Wesley Publishing Company, Reading, Massachusetts, USA, 1995).
-
5)
-
[1]. Pawlak, Z.: ‘Rough sets: theoretical aspects of reasoning about data’ (Kluwer Academic Publishers, Dordrecht, The Netherlands, 1991).
-
6)
-
[2]. Ali, R., Siddiqi, H.M., Lee, S.: ‘Rough set-based approaches for discretization: a compact review’, Artif. Intell. Rev., 2015, 44, (2), pp. 235–263 (doi: 10.1007/s10462-014-9426-2).
-
7)
-
[17]. Skowron, A., Stepaniuk, J.: ‘Tolerance approximation spaces’, Fundam. Inform., 1996, 27, pp. 245–253.
-
8)
-
[3]. Yang, Y., Webb, I. G., Wu, X.: ‘Discretization methods’, in Maimon, O., Rokachan, L. (Eds.): ‘Data mining and knowledge discovery handbook’ (Springer, Berlin, Germany, 2005), pp. 113–130.
-
9)
-
[21]. Imielinski, T., Lipski, W.: ‘Incomplete information in relational databases’, J. ACM, 1984, 31, pp. 761–791 (doi: 10.1145/1634.1886).
-
10)
-
[4]. Grzymala-Busse, J.W.: ‘Mining numerical data – a rough set approach’, Trans. Rough Sets, 2010, XI, pp. 1–13.
-
11)
-
[19]. Bosc, P., Duval, L., Pivert, O.: ‘An initial approach to the evaluation of possibilistic queries addressed to possibilistic databases’, Fuzzy Sets Syst., 2003, 140, pp. 151–166 (doi: 10.1016/S0165-0114(03)00032-0).
-
12)
-
[12]. Stefanowski, J., Tsoukiàs, A.: ‘Incomplete information tables and rough classification’, Comput. Intell., 2001, 17, pp. 545–566 (doi: 10.1111/0824-7935.00162).
-
13)
-
[14]. Nakata, M., Sakai, H.: ‘Twofold rough approximations under incomplete information’, Int. J. Gen. Syst., 2013, 42, pp. 546–571 (doi: 10.1080/17451000.2013.798898).
-
14)
-
[20]. Grahne, G.: ‘The problem of incomplete information in relational databases’, Lect. Notes Comput. Sci., 1991, 554 (doi: 10.1007/3-540-54919-6).
-
15)
-
[9]. Zenga, A., Lia, T., Liuc, D., et al: ‘A fuzzy rough set approach for incremental feature selection on hybrid information systems’, Fuzzy Sets Syst., 2015, 258, pp. 39–60 (doi: 10.1016/j.fss.2014.08.014).
-
16)
-
[15]. Lipski, W.: ‘On semantics issues connected with incomplete information databases’, ACM Trans. Database Syst., 1979, 4, pp. 262–296 (doi: 10.1145/320083.320088).
-
17)
-
[5]. Lin, T.Y.: ‘Neighborhood systems: a qualitative theory for fuzzy and rough sets’, Adv. Mach. Intell. Soft Comput., 1977, IV, pp. 132–155.
-
18)
-
[22]. Paredaens, J., De Bra, P., Gyssens, M., et al: ‘The structure of the relational database model’ (Springer-Verlag, Berlin, Germany, 1989).
-
19)
-
[8]. Yang, X., Zhang, M., Dou, H., et al: ‘Neighborhood systems-based rough sets in incomplete information system’, Inf. Sci., 2011, 24, pp. 858–867.
-
20)
-
[7]. Jing, S., She, K., Ali, S.: ‘A universal neighborhood rough sets model for knowledge discovering from incomplete hetergeneous data’, Expert Syst., 2013, 30, (1), pp. 89–96 (doi: 10.1111/j.1468-0394.2012.00633.x).
-
21)
-
[16]. Lipski, W.: ‘On databases with incomplete information’, J. ACM, 1981, 28, pp. 41–70 (doi: 10.1145/322234.322239).
-
22)
-
[10]. Zhao, B., Chen, X., Zeng, Q.: ‘Incomplete hybrid attributes reduction based on neighborhood granulation and approximation’. 2009 Int. Conf. on Mechatronics and Automation, 2009, pp. 2066–2071.
-
23)
-
[6]. Kryszkiewicz, M.: Rules in incomplete information systems. Inf. Sci., 1999, 113, pp. 271–292 (doi: 10.1016/S0020-0255(98)10065-8).
http://iet.metastore.ingenta.com/content/journals/10.1049/trit.2019.0025
Related content
content/journals/10.1049/trit.2019.0025
pub_keyword,iet_inspecKeyword,pub_concept
6
6