Your browser does not allow JavaScript!
JavaScript is necessary for the proper functioning of this website. Please enable JavaScript or use a modern browser.
Repository of the University of Ljubljana
Open Science Slovenia
Open Science
DiKUL
slv
|
eng
Search
Browse
New in RUL
About RUL
In numbers
Help
Sign in
Details
Ocenjevanje zanesljivosti regresijskih napovedi na podatkovnih tokovih
ID
HREN, BOŠTJAN
(
Author
),
ID
Kononenko, Igor
(
Mentor
)
More about this mentor...
PDF - Presentation file,
Download
(1,61 MB)
MD5: 9EFDB5D57855A29FD1F7AB83A6E61457
PID:
20.500.12556/rul/40de20a4-22b0-4d9d-956d-4bdd0d39ace8
Image galllery
Abstract
Z današnjo tehnologijo je mogoče preprosto neprekinjeno zbiranje podatkov. Kljub temu predstavlja pridobivanje znanja iz potencialno neskončnih podatkovnih tokov odprt problem. Zaradi določenih omejitev morajo biti metode za procesiranje podatkov dobro zasnovane, prostorsko učinkovite, računsko enostavne in hitre. Pogosto se analiza opravi na fiksni zgodovini podatkovnega toka, ki je določena z drsečim oknom. Kvaliteto napovedi algoritmov običajno ocenimo glede na njihovo povprečno točnost. Vendar, ko imamo opravka s podatki v realnem času, je lahko prav tako pomembna zanesljivost izhodnih vrednosti. V diplomskem delu obravnavamo ocenjevanje zanesljivosti posameznih napovedi pri učenju na podatkovnih tokovih. Raziskali smo različne metode, ki tvorijo intervalne ocene zanesljivosti s pristopom maksimalnega verjetja, stremljenja in lokalnih okolic, za delo na neprekinjenih dinamičnih podatkih. Metode smo implementirali z različnimi algoritmi strojnega učenja in jih preizkusili na več realnih in umetnih regresijskih problemih pri različnih velikostih drsečega okna. Uspešnost intervalnih cenilk smo ovrednotili z ocenami pokrivna verjetnost, relativni povprečni napovedni interval in kombinirana statistika. Primerjali smo izvajalne čase učnih algoritmov z in brez ocen zanesljivosti ter dosežene točnosti napovedi pri enakih časovnih omejitvah. Rezultate analiziramo tudi vizualno.
Language:
Slovenian
Keywords:
strojno učenje
,
ocenjevanje zanesljivosti
,
napovedni intervali
,
podatkovni tokovi
,
regresija
,
napovedovanje
Work type:
Undergraduate thesis
Organization:
FRI - Faculty of Computer and Information Science
Year:
2016
PID:
20.500.12556/RUL-84438
Publication date in RUL:
23.08.2016
Views:
1674
Downloads:
476
Metadata:
Cite this work
Plain text
BibTeX
EndNote XML
EndNote/Refer
RIS
ABNT
ACM Ref
AMA
APA
Chicago 17th Author-Date
Harvard
IEEE
ISO 690
MLA
Vancouver
:
HREN, BOŠTJAN, 2016,
Ocenjevanje zanesljivosti regresijskih napovedi na podatkovnih tokovih
[online]. Bachelor’s thesis. [Accessed 11 April 2025]. Retrieved from: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=eng&id=84438
Copy citation
Share:
Secondary language
Language:
English
Title:
Reliability estimation of regressional predictions on data streams
Abstract:
With today's technology it is easy to collect data continuously. Still, how to extract knowledge from potentially infinite data streams remains an open problem. Because of specific constraints, stream processing methods have to be well designed, space-efficient, computationally simple and fast. Typically, data analysis is done on a fixed history of the data stream defined by a sliding window. We usually define the quality of predictions by their average accuracy. However, when dealing with real-time data it can be also important to know the reliability of the models’ output values. In this thesis we deal with online reliability estimation of individual predictions on data streams. We consider different interval reliability estimators based on maximum likelihood, bootstrap and local neighborhood approach for working on continuous dynamic data. We implement these methods on different regression models and test them on several real and artificial regression problems with various sizes of the sliding window. Performance of the interval estimates are evaluated using the estimates of prediction interval coverage probability, the relative mean prediction interval and the combined statistic. We compare the execution times of learning algorithms with and without the reliability estimates as well as their prediction accuracy when given the same time constraint. We also analyze results visually.
Keywords:
machine learning
,
reliability estimation
,
prediction intervals
,
data stream
,
regression
,
predicting
Similar documents
Similar works from RUL:
Energy Turnover Dilemmas
Optimisation of quantity of electric energy from solar power plant regarding its layout
Prediction of daily photovoltaic systems production
Analysis of electric energy production in a small hydro power plant
Short-term forecast of electricity generation from photovoltaic and small-hydro power plants
Similar works from other Slovenian collections:
Investing in stocs
Automated trading strategies on the forex market
Preverjanje veljavnosti hipoteze učinkovitega trga: primer Stockholm Stock Exchange
Analiza gibanja vrednosti točke vzajemnih skladov slovenskih izdajateljev
Bubbles on financial markets
Back