Weighted Set Containment in Redundant Data Sets
Ladda ner
Typ
Examensarbete för masterexamen
Master Thesis
Master Thesis
Program
Computer science – algorithms, languages and logic (MPALG), MSc
Publicerad
2017
Författare
Grönvall, Johan
Wermensjö, Johan
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
Given a set family F and a query set Q, weighted set containment is the problem of finding the set C 2 F with the largest sum of weights, such that C Q, where every element has a corresponding weight. This problem was investigated at the request of Volvo Group IT, who rely heavily on weighted set containment queries in many of their applications. In this thesis we show that weighted set containment can be solved efficiently using trie based preprocessing when applied to redundant data sets. We show that finding the most efficient trie which represents F is NP-complete and we introduce a number of approximation algorithms. We show through empirical testing that some of our algorithms outperform state-of-the-art methods for similar problems when applied to Volvo’s particular data set.
Beskrivning
Ämne/nyckelord
Data- och informationsvetenskap , Computer and Information Science