Primerjalna analiza Evklidske in Poincaréjeve metrike v algoritmih strojnega učenja

magistrska naloga

Alenka Trpin (Author), Biljana Mileva-Boshkoska (Mentor)

Abstract

Živimo v času, ko si življenja brez računalnikov ne predstavljamo. Množična uporaba tako imenovane informacijsko komunikacijske tehnologije je proizvedla velike količine podatkov, ki jih sami ne moremo interpretirati in uporabiti. Z orodji podatkovnega rudarjenja in strojnega učenja se velike množice podatkov lahko obdelajo in uporabijo za napovedovanje in klasifikacijo. Eno od orodij za tako obdelavo podatkov je WEKA. Naloga temelji na osnovnem klasifikacijskem agoritem k najbližjih sosedov. V različnih panogah (gospodarstvo, zdravstvo, vojska...) se vedno bolj uporablja in shranjuje podatkovne baze raznovrstnih slik oziroma fotografij. Pri prepoznavanju podobosti med dvema fotografijama je pomembno, da algoritem prepozna določene vzorce. Prepoznavanje temelji na metriki. V ta namen je v orodje WEKA implementiran algoritem, ki temelji na Poincaréjevi metriki. Testiran je na podatkovni množici fotografij. Za namen primerjave je bil uporabljen algoritmom, ki temelji na evklidski metriki.

Keywords

podatkovno rudarjenje;strojno učenje;Poincaréjeva metrika;WEKA;k najbližjih sosedov;segmentacija;

Data

Language:	Slovenian
Year of publishing:	2018
Typology:	2.09 - Master's Thesis
Organization:	FIŠ - Faculty of Information Studies
Publisher:	[A. Trpin]
UDC:	004.85:004.421(043.2)
COBISS:	2048549907
Views:	816
Downloads:	63
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary abstract:	Today we cannot imagine life without computers. The massive use of the information communication technologies has produced large amounts of data that are difficult to interpret and use. With data mining tools and machine learning methods, large data sets can be processed and used for prediction and classification. One of the tools for such data processing is WEKA. The research in this thesis focuses on the basic classification algorithm the k nearest neighbors. In different industries (economy, health, military...) it increasingly uses and stores databases of various images or photographs. When recognizing the similarity between two photographs, it is important that the algorithm recognizes certain patterns. Recognition is based on metrics. For this purposes an algorithm based on Poincaré metric is implemented in WEKA and tested on a data set of photos. A comparison was made on algorithm based on Euclidean metric.
Secondary keywords:	data mining;machine learning;Poincaré metric;WEKA;k nearest neighbours;segmentation;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Fakulteta za informacijske študije v Novem mestu
Source comment:	Na ov.: Magistrska naloga : študijskega programa druge stopnje;
Pages:	IX, 67 str.
ID:	10990070