diplomsko delo
Domen Grzin (Author), Jurij Mihelič (Mentor)

Abstract

Predstavimo problem strnjenega polnjenja košev, podamo formalno definicijo problema in navedemo primer za boljšo predstavo bralcu ter kasnejšo razlago algoritmov. Predstavimo tudi preslikavo problema razvršča\-nja štud\-entov v predavalnice na ta problem in druge uporabe. Predstavimo tri natančne in dva približna algoritma, s katerimi rešujemo optimizacijske probleme, kot je problem strnjenega polnjenja košev. Razvijemo implementacije predstavljenih algoritmov za problem. Implementacije algoritmov ekperimentalno ovrednotimo in med seboj primerjamo po času, ki ga porabijo, da pridejo do rešitve. Približne algoritme primerjamo tudi po tem, kako blizu je njihova rešitev optimalni.

Keywords

optimizacijski problem;požrešni algoritem;izčrpno preiskovanje;razveji in omeji;iskanje v širino;eksperimentalno ovrednotenje;univerzitetni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [D. Grzin]
UDC: 004(043.2)
COBISS: 100742915 Link will open in a new window
Views: 187
Downloads: 26
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: Contiguous bin packing problem
Secondary abstract: We present the problem of contiguous bin packing, give a formal definition of the problem and an example, for a better presentation to the reader and an easier explanation of the algorithms in later chapters. We also present a mapping of classifying students into lecture halls onto our problem and other uses. We present three exact algorithms and two heuristic algorithms for solving optimization problems such as the problem of contiguous bin packing. We develop an implementation for each of the presented algorithms. After developing the implementations we experimentally evaluate them and compare with each other in terms of the time it takes for them to return a solution. For approximate algorithms we also compare their given solutions, specifically how close these are to the correct ones given by the exact algorithms.
Secondary keywords: optimization problem;greedy algorithm;exhaustive enumeration;branch and bound;breadth-first search;experimental evaluation;computer science;diploma;Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000468
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 44 str.
ID: 14728278