Υπολογισμός ανάστροφων ερωτημάτων κατάταξης σε κατανεμημένο περιβάλλον
View/
Keywords
Αλγόριθμοι ; Δεδομένα ; Reverse Top-k ; MapReduce ; HadoopAbstract
The last years has been proposed a new type of query, the Reverse Top-k query. This query has attracted the interest of both the research and the business community. This query can be found for example in business intelligence applications as could be used in order to predict the impact of a product to consumers. On the other hand, big data is an area with great activity, both from the business and the research area, as the data production from both business and individual is unprecedented. Thus in this thesis addressed the problem of processing Reverse Top-k query with parallel and distributed way. It is worth mentioning that this problem has not been tackled yet. Furthermore about the parallel and distributed process it used the MapReduce framework as is one of the most popular ways for such processing. In this thesis, initially two algorithms proposed to address this problem of which one proved better as achieves lower response times and can handle large volumes of data. Something resulting from the large number of experiments which performed during this work. The proposed algorithm is based on properties which are presented during this thesis and is a first attempt to address the problem of processing the Reverse Top-k query with parallel and distributed way.
Postgraduate Studies Programme
Ψηφιακά Συστήματα και ΥπηρεσίεςDepartment
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών ΣυστημάτωνNumber of pages
115Language
GreekCollections
Except where otherwise noted, this item's license is described as
Attribution-NonCommercial-NoDerivatives 4.0 Διεθνές
Attribution-NonCommercial-NoDerivatives 4.0 Διεθνές