1. Nature-inspired algorithms for hyperparameter optimizationFilip Glojnarić, 2019, magistrsko delo Opis: This master thesis is focusing on the utilization of nature-inspired algorithms for hyperparameter optimization, how they work and how to use them. We present some existing methods for hyperparameter optimization as well as propose a novel method that is based on six different nature-inspired algorithms: Firefly algorithm, Grey Wolf Optimizer, Particle Swarm Optimization, Genetic algorithm, Differential Evolution, and Hybrid Bat algorithm. We also show the optimization results (set of hyperparameters) for each algorithm and we present the plots of the accuracy for each combination and handpicked one. In discussion of the results, we provide the answers on our research questions as well as propose ideas for future work.
Ključne besede: artificial intelligence, artificial neural networks, machine learning, nature-inspired algorithms, evolutionary algorithms
Objavljeno: 09.12.2019; Ogledov: 829; Prenosov: 80
 Celotno besedilo (969,13 KB) |
2. FORMULATION, PREPARATION AND CHARACTERIZATION OF NANOEMULSIONS FOR PARENTERAL NUTRITIONDušica Mirković, 2019, doktorska disertacija Opis: The aim of this doctoral research was to develop and optimize parenteral nanoemulsions as well as the total parenteral nutrition (TPN) admixture containing a nanoemulsion obtained in the course of the optimization process (hereinafter referred to as optimal nanoemulsion), and to examine their physicochemical and biological quality as well. In addition, the quality of the prepared nanoemulsions was compared with the quality of the industrial nanoemulsion (Lipofundin® MCT/LCT 20%), and, in the end, the TPN admixture initially prepared was also compared with the admixture into which the industrial emulsion was incorporated.
Parenteral nanoemulsions that were considered in this dissertation were prepared by the high-pressure homogenization method. This method is the most widely applied method for the production of nanoemulsions due to the shortest length of homogenization time, the best-obtained homogeneity of the product and the smallest droplet diameter.
For the nanoemulsion formulation, preparation and optimization purposes, by using, firstly, the concept of the computer-generated fractional design, and, after that, the full experimental design, the assessment of both direct effects of different formulation and process parameters (the oil phase type, the emulsifier type and concentration, a number of homogenization cycles and the pressure under which homogenization was carried out) as well as the effects of their interactions on the characteristics of prepared nanoemulsions was performed.
Monitoring the nanoemulsion physical and chemical stability parameters was carried out immediately after their preparation, and then after 10, 30 and 60 days. It included the visual inspection, the measurement of the droplet diameter (the mean and volume droplet diameter), the polydispersity index, the ζ-potential, the pH value, the electrical conductivity, and the peroxide number. After the preparation and after 60 days, the biological evaluation (the sterility test and the endotoxic test) of the prepared nanoemulsions was carried out. As far as the characterization of the TPN admixture is concerned, it included practically the same parameters. The dynamics of monitoring the characteristics of the TPN admixture was determined on the basis of practical needs of hospitalized patients (0h, 24h and 72h).
The scope and comprehensiveness of this issue indicated the need to divide the doctoral dissertation into three basic stages. The first stage was preliminary. Using the 24-1 fractional factorial design, nanoemulsions for the parenteral nutrition were prepared. They contained either a combination of soybean and fish oil, or a combination of medium chain triglycerides and fish oil. In addition, the type and the amount of an emulsifier used, a number of high-pressure homogenization cycles, and the homogenization pressure, were also varied. The measurement of the above-mentioned parameters for the industrial nanoemulsion was parallely carried out (Lipofundin® MCT/LCT 20%). The objective of this part of the research was to identify critical numerical factors having the most significant effect on the characteristics that define the prepared parenteral nanoemulsions. Parameters that were singled out as the result of this stage of the research (the emulsifier concentration and a number of homogenization cycles) were used as independent variables in the second stage of the research.
Ključne besede: nanoemulsions, total parenteral nutrition admixtures, high pressure homogenization, design of experiments, optimization, analysis of variance, artificial neural networks
Objavljeno: 07.06.2019; Ogledov: 11011; Prenosov: 0
Celotno besedilo (2,82 MB) |
3. |
4. The accuracy of the germination rate of seeds based on image processing and artificial neural networksUroš Škrubej, Črtomir Rozman, Denis Stajnko, 2015, izvirni znanstveni članek Opis: This paper describes a computer vision system based on image processing and machine learning techniques which was implemented for automatic assessment of the tomato seed germination rate. The entire system was built using open source applications Image J, Weka and their public Java classes and linked by our specially developed code. After object detection, we applied artificial neural networks (ANN), which was able to correctly classify 95.44% of germinated seeds of tomato (Solanum lycopersicum L.).
Ključne besede: image processing, artificial neural networks, seeds, tomato
Objavljeno: 14.11.2017; Ogledov: 897; Prenosov: 301
Celotno besedilo (353,43 KB)
Gradivo ima več datotek! Več...
|
5. Analysis of neural network responses in calibration of microsimulation traffic modelIrena Ištoka Otković, Damir Varevac, Matjaž Šraml, 2015, izvirni znanstveni članek Opis: Microsimulation models are frequently used in traffic analysis. Various optimization methods are used in calibration, and the one method that has shown success is neural networks. This paper shows the responses of neural networks during calibration of a microsimulation traffic model. We analyzed two calibration methods by applying neural networks and comparing their neural network learning (according to their achieved correlation and the mean error of prediction) and their generalization ability (comparison of generalization results was analyzed in two steps). The best correlation between the microsimulation results and neural network prediction was 88.3%, achieved for the traveling time prediction, on which the first calibration method is based.
Ključne besede: microsimulation traffic models, calibration, response of neural networks, traveling time, queue parameters
Objavljeno: 02.08.2017; Ogledov: 745; Prenosov: 313
Celotno besedilo (1,07 MB)
Gradivo ima več datotek! Več...
|
6. Prediction of wine sensorial quality by routinely measured chemical propertiesAdriána Bednárová, Roman Kranvogl, Darinka Brodnjak-Vončina, Tjaša Jug, 2014, izvirni znanstveni članek Opis: The determination of the sensorial quality of wines is of great interest for wine consumers and producers since it declares the quality in most of the cases. The sensorial assays carried out by a group of experts are time-consuming and expensive especially when dealing with large batches of wines. Therefore, an attempt was made to assess the possibility of estimating the wine sensorial quality with using routinely measured chemical descriptors as predictors. For this purpose, 131 Slovenian red wine samples of different varieties and years of production were analysed and correlation and principal component analysis were applied to find inter-relations between the studied oenological descriptors. The method of artificial neural networks (ANNs) was utilised as the prediction tool for estimating overall sensorial quality of red wines. Each model was rigorously validated and sensitivity analysis was applied as a method for selecting the most important predictors. Consequently, acceptable results were obtained, when data representing only one year of production were included in the analysis. In this case, the coefficient of determination (R2) associated with training data was 0.95 and that for validation data was 0.90. When estimating sensorial quality in categorical form, 94 % and 85 % of correctly classified samples were achieved for training and validation subset, respectively.
Ključne besede: overall sensorial quality, prediction, Slovenian wine, artificial neural networks, multivariate data analysis
Objavljeno: 03.04.2017; Ogledov: 907; Prenosov: 322
Celotno besedilo (1,02 MB)
Gradivo ima več datotek! Več...
|
7. Novel Concepts for the Detection of MicroplasticsJan Ornik, 2016, magistrsko delo Opis: Microplastics are small pieces of plastic (smaller than 5 mm), which can be found in the environment and can be dangerous to living beings. It is expected that the abundance of microplastics will rise in the future. However, there are still no standard protocols for monitoring the microplastic abundance, which should include spectroscopic methods for an automated discrimination in order to produce reliable data. In this work we examined a new approach for microplastic detection based on the photoluminescence (PL) spectroscopy. To test the applicability of the proposed method a low-cost setup was built and characterized. The PL spectra from 27 different materials were collected and compared. The comparison of the spectra shows that the differentiation between samples is possible, especially between the plastic and non-plastic materials. Furthermore, the measured PL spectra also differ for different plastic types and other materials. However, the presence of dyes in plastic samples and incrustation of plastic samples by organic materials can affect the PL spectra and make the recognition troublesome. Disregarding organic materials and dyed plastic, the material differentiation based on the acquired PL spectra using neural networks resulted in 99.3 % accuracy when categorizing samples into plastic and non-plastic materials and 63.1 % accuracy when categorizing samples among different plastic and non-plastic materials. The promising results show that the PL spectroscopy of microplastics could outperform the spectroscopic methods used so far, by means of measurement speed and lateral resolution.
Ključne besede: microplastics, detection methods, photoluminescence spectroscopy, neural networks
Objavljeno: 10.08.2016; Ogledov: 1264; Prenosov: 125
Celotno besedilo (2,59 MB) |
8. Navzkrižno testiranje simboličnih in konektivističnih pristopov strojnemu učenju na specializiranih bazah akutnega vnetja slepičaMilan Zorman, Sandi Pohorec, Bojan Butolen, Bojan Žlahtič, Peter Kokol, 2012, izvirni znanstveni članek Ključne besede: acute appendicitis, decision trees, artificial neural networks, machine learning, cross-testing, medica informatics
Objavljeno: 30.12.2015; Ogledov: 982; Prenosov: 19
 Povezava na celotno besedilo |
9. Chemometric characterisation of the quality of ground waters from different wells in SloveniaErnest Vončina, Darinka Brodnjak-Vončina, Nataša Sovič, Marjana Novič, 2007, izvirni znanstveni članek Opis: The quality of ground water as a source of drinking water in Slovenia is regularly monitored. One of the monitoring programmes is performed on 5 wells for drinking water supply, 3 industrial wells and 2 ground water monitoring wells. Two hundred and fourteen samples of ground waters were analysed in the time 2003-2004. Samples were gathered from ten different sampling sites and physical chemical measurements were performed. The following 13 physical chemical parameters were regularly controlled: temperature, pH, conductivity, nitrate, AOX (adsorbable organic halogens), metals such as chromium, pesticides (desethyl atrazine, atrazine and 2,6-dichlorobenzamide), highly-volatile halogenated hydrocarbons (trichlorometane, 1,1,2,2-tetrachloroethene and 1,1,2-trichloroethene). For handling the results different chemometrics methods were employed, such as basic statistical methods for the determination of mean and median values, standard deviations, minimal and maximal values of measured parameters and their mutual correlation coefficients, cluster analysis (CA), the principal component analysis (PCA), the clustering method based on Kohonen neural network, and linear discriminant analysis (LDA). The study gives the opportunity to follow the quality of ground waters at different sampling sites within the defined time period. Monitoring of general pollution of ground waters and following measuring can be used to search the pollution source, to plan prevention measures and to protect from pollution, as well.
Ključne besede: ground waters, water quality, principal component analysis, classification, Kohonen neural networks
Objavljeno: 21.12.2015; Ogledov: 1492; Prenosov: 109
Celotno besedilo (390,30 KB)
Gradivo ima več datotek! Več...
|
10. Optimiranje značilnih parametrov frezanja z uporabo razvojne tehnike optimizacije jate delcevUroš Župerl, Franc Čuš, Valentina Gečevska, 2007, izvirni znanstveni članek Opis: Izbira rezalnih parametrov je najpomembnejši korak pri postopku načrtovanja proizvodnje, zato izdelamo novo tehniko razvojnega računanja za optimiranje procesa odrezovanja. V prispevku je uporabljena tehnika, ki oponaša dinamiko delcev v velikih skupinah (optimizacija PSO), za učinkovito in simultano optimiranje postopkov frezanja. V omenjenih postopkih smo soočeni s problemom več ciljnih dejavnikov. Najprej uporabimo umetno nevronsko mrežo (UNM) za napovedovanje rezalnih sil, nato z algoritmom PSO pridobimo optimalno rezalno hitrost in podajanja. Cilj optimizacije je, ob upoštevanju omejitev, določiti ekstrem ciljne funkcije (napovedna površina največjih sil). Med optimizacijo delci, s svojo inteligenco, letijo po prostoru rešitev in iščejo optimalne rezalne pogoje po strategiji algoritma PSO. Rezultati pokažejo, da je integriran sistem nevronskih mrež in kolektivne inteligence učinkovita metoda pri reševanju večciljnih optimizacijskih problemov. Njena velika učinkovitost na širokem območju rezalnih parametrov potrjuje, da sistem lahko praktično uporabimo v proizvodnji. Rezultati simulacij nakazujejo, da predlagan algoritem v primerjavi z rodovnimi algoritmi (GA) in simulacijskim (SA) ohlajanjem lahko poveča natančnost rešitve in konvergenco postopka. Nova tehnika razvojnega računanja ima nekoliko prednosti ter koristi in je primerna za uporabo v kombinaciji z modeli na osnovi umetnih nevronskih vezij, pri katerih niso na voljo izrecne relacije med vhodi in izhodi. Raziskava odpre vrata na področju obdelave z odrezovanjem za nov razred optimizacijskih tehnik, ki slonijo na razvojnem računanju.
Ključne besede: odrezovanje, končno frezanje, rezalni parametri, nevronske mreže, razvojne tehnike, optimizacija jate delcev, cutting, end-milling, cutting parameters, neural networks, evolution techniques, particle swarm optimization
Objavljeno: 10.07.2015; Ogledov: 987; Prenosov: 31
Povezava na celotno besedilo |
