magistrsko delo
Lovro Mažgon (Author), Vili Podgorelec (Mentor)

Abstract

V magistrskem delu je opisano področje priporočilnih sistemov. Zapisali smo njihovo formalno definicijo, naloge, ki jih opravljajo, ter vire podatkov in znanja. Navedli smo različne tehnike podajanja priporočil ter opisali pristope k uspešni evalvaciji priporočilnih sistemov. V praktičnem delu smo izdelali priporočilni sistem za priporočanje programskih ogrodij, namenjen programerjem. Implementirali smo tri priporočilne algoritme, ki bazirajo na podatkih v obliki grafa, in jih primerjali z algoritmom na osnovi asociacijskih pravil. Algoritme smo optimizirali in testirali na podatkih, pridobljenih s spletnega portala Stack Overflow. Pridobljeni rezultati nakazujejo, da ima uporabljen pristop visok potencial ter da je v pravih razmerah smiseln in uporaben.

Keywords

priporočilni sistemi;grafi;naključni sprehodi;asociacijski pravila;magistrske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher: L. Mažgon
UDC: 004.4'275:004.021(043.2)
COBISS: 19796758 Link will open in a new window
Views: 885
Downloads: 158
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Implementation of a Recommender System Based on a Graph Extracted from Itemsets
Secondary abstract: The thesis addresses the area of recommender systems. We described their formal definition, tasks, data and knowledge sources. Furthermore, we defined various recommender techniques as well as approaches to the successful evaluation of recommender systems. Our empirical work covers the construction of a recommender system for programmers, which recommends frameworks. We implemented three recommender algorithms, which are based on graph data, as well as one based on association rules. We optimized the algorithms and tested them on data obtained from the website Stack Overflow. The results suggest that the approach has high potential and that it is reasonable and useful in the right circumstances.
Secondary keywords: recommender system;graph;randon walk;associaton rules;
URN: URN:SI:UM:
Type (COBISS): Master's thesis/paper
Thesis comment: Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages: VIII, 75 str.
ID: 9147230
Recommended works: