| | SLO | ENG | Piškotki in zasebnost

Večja pisava | Manjša pisava

Iskanje po katalogu digitalne knjižnice Pomoč

Iskalni niz: išči po
išči po
išči po
išči po
* po starem in bolonjskem študiju

Opcije:
  Ponastavi


1 - 10 / 10
Na začetekNa prejšnjo stran1Na naslednjo stranNa konec
1.
2.
Agile Machine Learning Model Development Using Data Canyons in Medicine : A Step towards Explainable Artificial Intelligence and Flexible Expert-Based Model Improvement
Bojan Žlahtič, Jernej Završnik, Helena Blažun Vošner, Peter Kokol, David Šuran, Tadej Završnik, 2023, izvirni znanstveni članek

Opis: Over the past few decades, machine learning has emerged as a valuable tool in the field of medicine, driven by the accumulation of vast amounts of medical data and the imperative to harness this data for the betterment of humanity. However, many of the prevailing machine learning algorithms in use today are characterized as black-box models, lacking transparency in their decision-making processes and are often devoid of clear visualization capabilities. The transparency of these machine learning models impedes medical experts from effectively leveraging them due to the high-stakes nature of their decisions. Consequently, the need for explainable artificial intelligence (XAI) that aims to address the demand for transparency in the decision-making mechanisms of black-box algorithms has arisen. Alternatively, employing white-box algorithms can empower medical experts by allowing them to contribute their knowledge to the decision-making process and obtain a clear and transparent output. This approach offers an opportunity to personalize machine learning models through an agile process. A novel white-box machine learning algorithm known as Data canyons was employed as a transparent and robust foundation for the proposed solution. By providing medical experts with a web framework where their expertise is transferred to a machine learning model and enabling the utilization of this process in an agile manner, a symbiotic relationship is fostered between the domains of medical expertise and machine learning. The flexibility to manipulate the output machine learning model and visually validate it, even without expertise in machine learning, establishes a crucial link between these two expert domains.
Ključne besede: XAI, explainable artificial intelligence, data canyons, machine learning, transparency, agile development, white-box model
Objavljeno v DKUM: 14.03.2024; Ogledov: 299; Prenosov: 30
.pdf Celotno besedilo (5,28 MB)
Gradivo ima več datotek! Več...

3.
Stohastično modeliranje obrestnih mer
Ines Štampar, 2020, magistrsko delo

Opis: Magistrsko delo obravnava napoved obrestnih mer in vpliv gibanja obrestnih mer na anuiteto dolgoročnega kredita. V prvem delu je na kratko povzeta teorija stohastičnih procesov, Brownovega gibanja in Itôvega procesa. Za napoved obrestnih mer so bili uporabljeni Vasickov, CIR in Hull-Whiteov model. V drugem delu so opisane lastnosti modelov ter izpeljava pričakovane vrednosti in variance. V tretjem delu sledi modeliranje 3-mesečnega Euribor-ja. Uporabljena je metoda največje verjetnosti za Vasickov in CIR model, za Hull-Whiteov model pa metoda najmanjšega verjetja. Vključene so napovedi posameznega modela in pregled gibanja naslednjih 20 let. V četrtem delu so analizirani možni načini najema dolgoročnega kredita, predvsem odločitev o fiksni ali spremenljivi obrestni meri. Glede na dobljene rezultate napovedi obrestnih mer je sestavljen amortizacijski načrt in potek dolgoročnega kredita. Delo je zaključeno s poglavjem, kjer so podani odgovori na vprašanje, ali se splača najeti nov kredit in poplačati starega (glede na nizke vrednosti trenutnih obrestnih mer).
Ključne besede: Stohastični model, obrestne mere, Vasicek, CIR, Hull-White, napoved, kredit, amortizacija
Objavljeno v DKUM: 20.01.2021; Ogledov: 1143; Prenosov: 111
.pdf Celotno besedilo (736,59 KB)

4.
Obravnava pacienta s sindromom Wolff-Parkinson-White
Mark Duh, 2020, diplomsko delo

Opis: Uvod: Značilnost sindroma Wolff-Parkinson-White so dodatne poti v srcu pacientov, ki vodijo impulze od atrija do prekata. Najpogostejši posledici sta hiter in nereden pulz, ki se pogosteje pojavljata v obliki napadov. Namen zaključnega dela je raziskati sindrom Wolff-Parkinson-White, kakšne so naloge medicinske sestre pri obravnavi pacientov s sindromom Wolff-Parkinson-White, s kakšnimi težavami se pacient s sindromom Wolff-Parkinson-White sooča, kako se nanje odziva ter kakšna je njegova kakovost življenja pri opravljanju telesne dejavnosti. Metode: V teoretičnem delu je bila uporabljena deskriptivna metoda dela, v empiričnem delu kvalitativna metodologija, kjer smo izvedli študijo primera pacienta s sindromom Wolff-Parkinson-White s pomočjo delno strukturiranega intervjuja. Rezultati: Najpogostejše težave pacienta, so hiter in nereden srčni utrip, nemoč za nadaljevanje telesne aktivnosti, težje dihanje ob večjem naporu ter utrujenost po napadu simptomatike. Težave se zmanjšajo ob prenehanju aktivnosti, ali ob izvajanju dihalnih vaj za sproščanje. Kakovost življenja pri izvajanju telesne aktivnost je pri pacientu s sindromom Wolff-Parkinson-White na dokaj nizki ravni in je še nižja, kadar pacient ne upošteva navodil, ki mu jih je predpisal kardiolog. Razprava in sklep: Sindrom Wolff-Parkinson-White je dokaj nepoznan sindrom. Težave, ki se pri njem pojavljajo, znižujejo kakovost življenja pacientov in puščajo negativne posledice. Še posebej se to kaže pri pacientih, ki se aktivno ukvarjajo s tekmovalnim športom. Ključne naloge medicinskih sester pri obravnavi pacientov so usmerjene v preventivno zdravstveno vzgojo ter obravnavo pred in po radio-frekvenčni katetrski ablaciji.
Ključne besede: Wolff-Parkinson-White, ablacija, kakovost življenja, zdravstvena nega, telesna aktivnost
Objavljeno v DKUM: 21.09.2020; Ogledov: 1538; Prenosov: 208
.pdf Celotno besedilo (456,02 KB)

5.
Public and private policing of financial crimes : the struggle for co-ordination
Michael Levi, 2010, pregledni znanstveni članek

Opis: Purpose: The purpose of doing this is to move our attention, both theoretically and empirically, away from looking solely at the police as deliverers of “policing” and more in the direction of the web of police, non-police agencies (like financial regulators and departmental investigators in the UK), global and local private interveners (like payment card systems, banks, and insurance companies). Design/Methods/Approach: Interviews with high officials and operational public and private sector personnel, principally in the UK but also in continental Europe, North America and Australasia. Findings: This paper deals with the evolving shape of the control of financial crimes, especially frauds. Most policing activity in the area of UK public sector fraud is undertaken not by the police but by departmental investigators; most frauds against business come to the attention of the police via corporate actors; while frauds against the public may be reported direct, or via intermediaries like credit card issuers. It is difficult for public sector bodies to escape their narrow departmental orientations in developing “joint working” in public-public partnerships. Publicprivate partnerships have been developing rapidly in dealing with volume frauds such as payment card and insurance frauds, but less rapidly in more elite whitecollar crime areas such as management fraud, which are less predictable and where elites resist accountability. Practical implications: It is misconceived to look only at public police activity when assessing the scope and impact of fraud control. The research shows that considerable efforts are made by the police and different parts of the public and the private sector in the UK to harmonise intelligence, but scarcity of investigative resources means that both the private and public sectors are frustrated in getting action on their intelligence product.
Ključne besede: fraud, white-collar crime, economic crime, public-private partnerships, policing
Objavljeno v DKUM: 12.05.2020; Ogledov: 886; Prenosov: 49
URL Povezava na celotno besedilo
Gradivo ima več datotek! Več...

6.
The Influence of various origins of first calving Simmental and Black-White cows on production and content of milk
Marjan Janžekovič, Dejan Škorjanc, Jože Smolinger, 2004, izvirni znanstveni članek

Opis: A total of 1 070 Simmental and Black-White first calving cows and the same number of mother cows were investigated. The first calving cows originally arrived from Slovenia, the Netherlands, Germany and Denmark. No statistically significant differences were established in the milk production and quality of milk between the first calving cows from abroad compared to the first calving Slovenian cows. Statistically significant higher milk production (8 004 ± 981 kg; P ≤0.001) had mother cows of the imported Black-White compared to the Slovene (6 514 ± 1 142 kg) ones. There were no statistically significant differences, in the milk production and quality of milk, between the imported mothers of first calving cows of Simmental breed comparing with Slovenian mothers of first calving cows of the same breed. The milk production between the mothers and their Simmental daughters had a statistically significant correlation (rp = 0.125; P ≤ 0.01). The higher correlation coefficient for this characteristic was established between the Black-White mothers and their daughters (rp = 0.302; P ≤ 0.001). Among the milk production of first calving cows and protein content in milk a statistically negative significant correlation (rp = -0.198; P ≤ 0.001) was established. This result confirmed that the purchase of more expensive animals with better genetic potential is economically unjustified for the breeder who does not reach intensive technology of basic quality voluminous fodder.
Ključne besede: cattle, Simmental, Black-White, milk production, protein, fat
Objavljeno v DKUM: 10.07.2017; Ogledov: 1474; Prenosov: 105
.pdf Celotno besedilo (225,07 KB)
Gradivo ima več datotek! Več...

7.
ERROR PROBABILITY MODEL FOR IEEE 802.15.4 WIRELESS TRANSMISSION WITH CO-CHANNEL INTERFERENCE AND BACKGROUND NOISE
Uroš Pešović, 2016, doktorska disertacija

Opis: Data transmission sent through wireless channel is usually affected by background noise, multipath fading and interference which cause data errors. Influence of such disturbances is the most commonly expressed in a form of error probability statistics. Effects of these disturbances on IEEE 802.15.4 wireless transmissions are previously studied, except influence of co-channel interference (CCI) which originates from collision between IEEE 802.15.4 devices which perform simultaneous radio transmission. Our thesis puts forward the assumption that it is possible to derive more accurate analytical error probability model for higher data level error probability parameters without the idealization of PN spreading sequences. Additionally, thesis is that is possible to derive an accurate analytical error probability model in the case of CCI influenced by background noise by consideration of constellation diagram. IEEE 802.15.4 standard uses CSMA/CA (Carrier Sense Multiple Access with Collision Detection) channel access mechanism to prevent collisions between devices, but this mechanism doesn't provide protection from hidden node problem which is primary source of co-channel interference. Using Monte Carlo simulations we determined frequency of hidden node collisions, which shown that co-channel interference frequently occur in parts of the network with high traffic load. Some prior works in this field tend to idealize these non-ideal spreading sequences in order to simplify calculations for error probability parameters. Our doctor thesis presents analytical model of data level error probability parameters (symbol, bit and packet) for IEEE 802.15.4, which uses original non-ideal spreading sequences without their idealization. Proposed error probability model consists of mutually dependent chip, symbol, bit and packet error probability models. Derived error probability models are linked together, so each of error probability parameters can be determined using error probability parameter from the previous stage. Error probability model for IEEE 802.15.4 wireless communication could be used in network simulation tools in order to accurately simulate energy efficient medium access protocols in realistic scenarios. Presented theoretical results are tested by independent numerical simulation of IEEE 802.15.4 transmission according to Monte Carlo method. Simulation results shows that derived models for error probability parameters were matched by two simulation scenarios in background noise, for multipath fading and co-channel interface, respectively Furthermore, the accuracy of derived mathematical models was tested in real-world experiment using IEEE 802.15.4 compliant wireless transceivers for creating co-channel interference. Packets were received by software defined radio platform, which enabled realization of coherent receiver in which all error probability statistics could be collected. The results of the experiment show consistency with proposed analytical error probability models, but some deviations are caused by poor preamble synchronization under low SNR (Signal to Noise Ratio) value. The thesis was proved with Monte Carlo simulations of the physical level of the IEEE 802.15.4 communication and experimental measurements on a real physical communication system.
Ključne besede: IEEE 802.15.4 standard, error probability model, co-channel interference, Rician fading channel, additive white Gaussian noise, wireless transmission, wireless sensor networks, numerical simulations, software defined radio
Objavljeno v DKUM: 14.10.2016; Ogledov: 2502; Prenosov: 140
.pdf Celotno besedilo (8,20 MB)

8.
9.
Concept and design of electric car body
Marko Sovič, 2012, diplomsko delo

Opis: In this thesis, a concept multi-material body for an electric vehicle is presented. A selection criteria for materials is made with regard to component function, their traditional shape and the mode of load they have to carry. Initial component design is based on a vehicle reference model. Numerous component modifications are done in dependence on the chosen materials and their production processes. Special attention is payed to the electric vehicle's body structural changes that need to be performed due to the passenger and battery safety. A program for distance simulation is made and used to evaluate cost changes, on a whole vehicle scale, when performing mass-reduction in a vehicle body. It has been proven that it is justified to use light materials.
Ključne besede: body-in-white, electric vehicle, material selection, Catia V5, distance simulation, cost analysis.
Objavljeno v DKUM: 07.03.2013; Ogledov: 1834; Prenosov: 192
.pdf Celotno besedilo (7,79 MB)

10.
GRAPE ANTIOXIDANTS - INFLUENCE OF PROCESSING PARAMETERS ON THEIR CONTENT AND ACTIVITY
Iva Ružić, 2012, doktorska disertacija

Opis: Grapes and wine production residuals, containing grape skins, seeds and stems, represent important sources of phenolic compounds. Red wines are rich in phenolic antioxidants while white wines generally show lower (poly)phenolic content and antioxidant activity. The main difference in the production of white and red wines is the fermentation step. The fermentation of red grapes includes maceration, i.e. it takes place in both liquid and solid parts of the grape. In the case of white wines, the must is clarified before the fermentation starts, cultured yeasts are added to convert the must to wine and antioxidant sulfur dioxide is inserted to protect the wine. The objective of the PhD research was to study antioxidant potential of white grape phenolics both in raw material and fermentation products. First, the standards of phenolic compounds commonly found in white grapes and wines were tested for their antioxidant and antiradical activity. All tested compounds except apigenin showed strong antioxidant and antiradical properties. Afterwards, conventional solid-liquid extraction (CE) and pressurized liquid extraction (PLE) of phenolic compounds from grape seeds were performed. The optimal extraction conditions were investigated and extracts were analyzed in order to determine their total phenolic content, individual phenolic compounds and antiradical potential. Two different drying methods (hot air and freeze drying) were applied to the seeds. Solvent efficiencies of water, ethanol and water-ethanol mixtures were compared at different temperatures. In addition, supercritical fluid extraction (SFE) with CO2 was investigated as the technique for the recovery of phenolic compounds and for the pretreatment of white grape seed samples. Both CE and PLE gave extracts rich in phenolic compounds with good antiradical properties and, generally, the best results were obtained with PLE at 130°C. At the end, the influence of maceration process on the content and the activity of grape phenolic antioxidants in white wines was studied. For such a purpose, six commercially available macerated white wines from Italy and Croatia were analyzed. In addition, eight non-macerated wines from the same region and variety were studied and compared to macerated wines. The experiments have shown that the maceration process increases the total phenolic content and antiradical activity of the wine. It enables the production of white wines rich in phenolic compounds and with strong antiradical properties in completely natural way using only the grapes and without addition of any chemical agents.
Ključne besede: white grapes, white wine, maceration, antioxidant activity, polyphenols, extraction, PLE, SFE.
Objavljeno v DKUM: 05.04.2012; Ogledov: 2968; Prenosov: 182
.pdf Celotno besedilo (5,65 MB)

Iskanje izvedeno v 0.24 sek.
Na vrh
Logotipi partnerjev Univerza v Mariboru Univerza v Ljubljani Univerza na Primorskem Univerza v Novi Gorici