Finding skyline points over dynamic incomplete database

• Main Authors: Babanejad, Ghazaleh, Ibrahim, Hamidah, Udzir, Nur Izura, Sidi, Fatimah, Aljuboori, Ali A.Alwan
• Format: Conference or Workshop Item
• Language: English
• Published: Faculty of Computer Science and Information Technology, Universiti Putra Malaysia 2014
• Subjects: T Technology (General)
• Online Access: http://irep.iium.edu.my/40151/1/Finding_Skyline_Points_over_Dynmic_Incomplete_Database.pdf; http://irep.iium.edu.my/40151/; http://mancod2014.blogspot.com/p/proceedings.html

In database systems with growing data, finding results that best meet the user’s preferences is important. Skyline queries play an essential role, in which only the data items from the database that are not being dominated by the other data items are returned as results. Most of the existing skyline approaches assume that the databases are complete which means there are no missing values in the database dimensions. However, in reality databases are not complete particularly for multidimensional database in which some dimensions have missing values. Missing values change the native of dominance relation and have a negative effect on skyline technique which leads to unsatisfying the transitivity property of skylines and incur cyclic dominance. This problem becomes more complicated when the database is dynamic in which new items are inserted into the database while some items are deleted or updated. Moreover, most of the works that consider incomplete database are over static items. This paper attempts to propose a new approach that identifies the most relevant items that best meet the user’s preferences in incomplete and dynamic databases.