1. |
2. Rekurzivna delitev modelov linearne regresije za oceno zanimivosti asociativnih pravil v različnih časovnih obdobjihGoran Hrovat, 2018, doktorska disertacija Opis: Zanimivosti asociativnih pravil ali pogostih množic postavk se lahko skozi čas spreminjajo. Prav tako je lahko njihova zanimivost različna za različne skupine (npr. skupine ljudi). V doktorski disertaciji je predstavljen nov algoritem za določanje zanimivosti asociativnih pravil in množic postavk v različnih časov¬nih obdobjih. Algoritem odkriva skupine pacientov, ki se glede na trend zanimivosti asociativnega pravila statistično značilno razlikujejo. Rezultat algoritma je odločitveno regresijsko drevo, katerega povezave predstavljajo različne skupine pacientov, listi pa prikazujejo grafe z zanimivostmi asociativnega pravila ali množice postavk v različnih časovnih obdobjih. Pokazali smo tudi, da podpora pogoste množice postavk vpliva na kompleksnost zgrajenega regresijskega drevesa. Za demonstracijo smo uporabili elektronske zdravstvene zapise, zbrane v obdobju 15 let, ki predstavljajo odpuste pacientov iz različnih bolnišnic v Združenih državah Amerike. Predstavljeni algoritem predstavlja v tem primeru uporabno vrednost za bolnišnice, zavarovalnice, raziskovalne in ostale institucije, saj jih lahko odkrito znanje vodi do novih spoznanj in optimizacije poslovanja.
Ključne besede: podatkovno rudarjenje, mere zanimivosti, asociativna pravila, podpora odločanju, regresijsko dre-vo, linearna regresija, elektronski zdravstveni zapis
Objavljeno v DKUM: 12.07.2018; Ogledov: 2215; Prenosov: 150
 Celotno besedilo (17,46 MB) |
3. Hybrid Web Application Development Using Meteor.js FrameworkNejc Burič Novačan, 2016, diplomsko delo Opis: This Thesis work shows one of many possible ways of developing hybrid web applications. In the theoretical part, we described the Meteor.js framework and MongoDB database, which we used for realization of the web application. In the presentation of the framework, new and modern methods are described which are used in modern web programming. The hybrid web application which we developed in the practical part enables the user to navigate to tourist attractions in Slovenia. It enables the user to discover and locate natural landmarks, rural tourism, clubs, electric car charging stations, places of interest,, and Tourist Information Centres. Admin can regulate the points of interest with the help of Google API. The application also displays current distances to every point of interest and notifies the users about the changes made by the administrator.
Ključne besede: web programming, hybrid applications, Meteor.js, MongoDB
Objavljeno v DKUM: 22.09.2016; Ogledov: 2427; Prenosov: 157
Celotno besedilo (2,48 MB) |
4. Projektna dokumentacija projekta ODUNMilan Ojsteršek, Janez Brezovnik, Mojca Kotar, Nelson Šorgo, Petruša Miholič, Vanesa Valentinčič, Bojan Korečič, Zoran Mesec, Anton Jagodic, Marko Ferme, Mladen Borovič, Albin Bregant, Goran Hrovat, 2013, projektna dokumentacija (idejni projekt, izvedbeni projekt) Ključne besede: projektna dokumentacija, ODUN
Objavljeno v DKUM: 15.07.2015; Ogledov: 4183; Prenosov: 118
 Povezava na celotno besedilo |
5. Constructing domain-specific semantic dictionaries to supplement domain-specific knowledge basesGoran Hrovat, Milan Ojsteršek, objavljeni znanstveni prispevek na konferenci Opis: Semantic dictionaries as well as knowledge bases are im- portant source of information for natural language process- ing. Using corpus and algorithms for constructing semantic space, we can quickly construct semantic dictionary, which is exploit to supplement the knowledge base. Algorithm for constructing semantic space, COALS was chosen. Semantic distance between terms in semantic space reveals their sim- ilarity, which is used to add semantic relationships in the dictionary. Semantic dictionary also serves to assist in iden- tifying entities and in relation extraction, to supplement the knowledge base.
Ključne besede: semantic space, correlation, semantic dictionary, knowledge base
Objavljeno v DKUM: 22.06.2015; Ogledov: 1937; Prenosov: 46
Povezava na datoteko |
6. Vzpostavitev repozitorijev slovenskih univerz in nacionalnega portala odprte znanostiMilan Ojsteršek, Mojca Kotar, Marko Ferme, Goran Hrovat, Mladen Borovič, Albin Bregant, Jan Bezget, Janez Brezovnik, 2014, izvirni znanstveni članek Opis: V prispevku so predstavljeni pravni, organizacijski in tehnični vidiki vzpostavitve slovenske nacionalne infrastrukture odprtega dostopa do zaključnih del študija ter objav raziskovalcev, ki je sestavljena iz štirih institucionalnih repozitorijev slovenskih univerz in nacionalnega portala. Slednji agregira vsebine iz univerzitetnih repozitorijev in drugih slovenskih zbirk za potrebe skupnega iskalnika, priporočilnega sistema in detektorja podobnih vsebin. Razvita je bila programska oprema, ki je povezana z informacijskimi in avtentikacijskimi sistemi univerz ter s sistemom COBISS.SI. V okviru projekta so bile definirane pravne podlage, implementirani procesi za oddajo zaključnih del študija in objav zaposlenih, proces prenosa podatkov v nacionalni portal ter procesi izmenjave podatkov o priporočanju sorodnih del in primerjave podobnih vsebin. Uporabnikom z vsega sveta so na voljo dvojezične spletne in mobilne aplikacije, priporočilni sistem ter funkcionalnosti za uporabnike s posebnimi potrebami. Repozitoriji so učinkovito promocijsko orodje za univerze in njihove raziskovalce. Pričakovati je, da bodo izboljšali prepoznavnost slovenskih univerz v svetu. Kompleksno zasnovana nacionalna infrastruktura odprtega dostopa z detektorjem podobnih vsebin in s povezanimi zunanjimi sistemi bo omogočila hranjenje skoraj osemdesetih odstotkov recenziranih znanstvenih del, ki jih na leto objavijo slovenski raziskovalci. Prav tako bo dostopna večina zaključnih del študija, ki vsako leto nastanejo na slovenskih visokošolskih zavodih.
Ključne besede: odprti dostop, nacionalne infrastrukture, institucionalni repozitoriji, preverjanje podobnosti vsebin, priporočilni sistemi
Objavljeno v DKUM: 21.12.2014; Ogledov: 1783; Prenosov: 150 (3 glasovi)
Celotno besedilo (784,34 KB)
Gradivo ima več datotek! Več... |
7. Establishing of a Slovenian open access infrastructure: a technical point of viewMilan Ojsteršek, Janez Brezovnik, Mojca Kotar, Marko Ferme, Goran Hrovat, Albin Bregant, Mladen Borovič, 2014, izvirni znanstveni članek Opis: Purpose – This paper presents a technical perspective when implementing the Slovenian open access infrastructure that consists of four institutional repositories and a national portal that aggregates content from the repositories in order to provide a common search engine, recommendations of similar documents, and similar text detection.
Design/methodology/approach – During the project, the necessary legal background and processes for mandatory submissions of final study works, research publications and research data were established, as well as processes for data exchange between the institutional repositories and the national portal, and processes for similar text detection.
Findings – The consortium consisted of four Slovenian universities that significantly differ in size, organisation, and workflows. It was anticipated that exactly the same legal background and software would be used for the four repositories. It turned out that complete unification was impossible due to the differences.
Practical implication – The national open access infrastructure will improve the visibility of Slovenian research organisations. It supports the compliance with the funders’ open access mandates. The established infrastructure enables the depositing and archiving of approximately eighty percent of the peer-reviewed scientific publications that are annually published by Slovenian researchers. At the same time, the majority of final study works from Slovenian higher education institutions are available in full-text format.
Originality/value – This paper describes a technical perspective for setting up a national open access infrastructure, which has not been described in the literature previously.
Ključne besede: institutional repositories, open access, national open access infrastrukture, plagiarism detection, recommendation system
Objavljeno v DKUM: 27.11.2014; Ogledov: 3051; Prenosov: 609 (2 glasa)
Celotno besedilo (981,64 KB)
Gradivo ima več datotek! Več... |
8. Contrasting temporal trend discovery for large healthcare databasesGoran Hrovat, Gregor Štiglic, Peter Kokol, Milan Ojsteršek, izvirni znanstveni članek Opis: With the increased acceptance of electronic health records, we can observe theincreasing interest in the application of data mining approaches within this field. This study introduces a novel approach for exploring and comparingtemporal trends within different in-patient subgroups, which is basedon associated rule mining using Apriori algorithm and linear model-based recursive partitioning. The Nationwide Inpatient Sample (NIS), Healthcare Costand Utilization Project (HCUP), Agency for Healthcare Research and Qualitywas used to evaluate the proposed approach. This study presents a novelapproach where visual analytics on big data is used for trend discovery in form of a regression tree with scatter plots in the leaves of the tree. Thetrend lines are used for directly comparing linear trends within a specified time frame. Our results demonstrate the existence of opposite trendsin relation to age and sex based subgroups that would be impossible to discover using traditional trend-tracking techniques. Such an approach can be employed regarding decision support applications for policy makers when organizing campaigns or by hospital management for observing trends that cannot be directly discovered using traditional analytical techniques.
Ključne besede: data mining, decision support, trend discovery
Objavljeno v DKUM: 27.11.2014; Ogledov: 2025; Prenosov: 664
Celotno besedilo (1013,97 KB)
Gradivo ima več datotek! Več... |
9. |
