352 research outputs found
Large-scale test data set for location problems
No full textDesigners of location algorithms share test data sets (benchmarks) to be able to compare performance of newly developed algorithms. In previous decades, the availability of locational data was limited. Big data has revolutionised the amount and detail of information available about human activities and the environment. It is expected that integration of big data into location analysis will increase the resolution and precision of input data. Consequently, the size of solved problems will significantly increase the demand on the development of algorithms that will be able to solve such problems. Accessibility of realistic large scale test data sets, with the number of demands points above 100,000, is very limited. The presented data set covers entire area of Slovakia and consists of the graph of the road network and almost 700,000 connected demand points. The population of 5.5 million inhabitants is allocated to the locations of demand points considering the residential population grid to estimate the size of the demand. The resolution of demand point locations is 100 m. With this article the test data is made publicly available to enable other researches to investigate their algorithms. The second area of its utilisation is the design of methods to eliminate aggregation errors that are usually present when considering location problems of such size. The data set is related to two research articles: “A Versatile Adaptive Aggregation Framework for Spatially Large Discrete Location-Allocation Problem” (Cebecauer and Buzna, 2017) [1] and “Effects of demand estimates on the evaluation and optimality of service centre locations” (Cebecauer et al., 2016) [2]
Short-Term Traffic Prediction in Large-Scale Urban Networks
No full textCity-wide travel time prediction in real-time is an important enabler for efficient use of the road network. It can be used in traveler information to enable more efficient routing of individual vehicles as well as decision support for traffic management applications such as directed information campaigns or incident management. 3D speed maps have been shown to be a promising methodology for revealing day-to-day regularities of city-level travel times and possibly also for short-term prediction. In this paper, we aim to further evaluate and benchmark the use of 3D speed maps for short-term travel time prediction and to enable scenario-based evaluation of traffic management actions we also evaluate the framework for traffic flow prediction. The 3D speed map methodology is adapted to short-term prediction and benchmarked against historical mean as well as against Probabilistic Principal Component Analysis (PPCA). The benchmarking and analysis are made using one year of travel time and traffic flow data for the city of Stockholm, Sweden. The result of the case study shows very promising results of the 3D speed map methodology for short-term prediction of both travel times and traffic flows. The modified version of the 3D speed map prediction outperforms the historical mean prediction as well as the PPCA method. Further work includes an extended evaluation of the method for different conditions in terms of underlying sensor infrastructure, preprocessing and spatio-temporal aggregation as well as benchmarking against other prediction methods.QC 20190531</p
Enhancing Short-Term Traffic Prediction for Large-Scale Transport Networks by Spatio-Temporal Clustering
No full textCongestion in large cities is responsible for extra travel time, noise, air pollution, CO2 emissions, and more. Transport is one of the main recognized contributors to global warming and climate change, which is getting increasing attention from authorities and societies around the world. Better utilization of existing resources by Intelligent Transport Systems (ITS) and digital technologies are recognized by the European Commission as technologies with enormous potential to lower the negative impacts associated with high traffic volumes in urban areas. The main focus of this work is on short-term traffic prediction, which is an essential tool in ITS. In combination with providing information, it enables proactive decisions to decrease severity of congestion that occurs regularly or is caused by incidents. The main contribution of this work is to develop a methodological framework and prove its enhancing effects on short-term prediction in the context of large-scale transport networks. It is expected to contribute to more robust and accurate predictions of ITS in traffic management centers. Traffic patterns in large-scale networks, including urban streets, can be heterogeneous during the day and from day-to-day. This work investigates spatio-temporal clustering of heterogeneous data sets to smaller, more homogeneous data sub-sets. This is expected to produce more robust, accurate, scalable, and cost-effective prediction models. This thesis is the collection of five papers that contribute to enhancing short-term traffic prediction in this context. The clustering is recognized to boost prediction performance in Papers II, III, IV, and V. Paper II considers network partitioning and the last three papers study day clustering. The prediction models used across included papers are naive historical mean prediction models and more advanced prediction models such as probabilistic principal component analysis (PPCA) and exponential smoothing. Paper I considers and facilitates floating car data (FCD) as a cost-effective opportunistic source of speed and travel time data with extensive network coverage. Common practice in determining the number of clusters is to rely on internal evaluation indices, and these are very efficient but isolated from application. Paper IV tests this practice by also considering performance in short-term prediction application. Our results show that relying on these indices can lead to a loss of prediction accuracy of about 20% depending on the considered prediction model. Dimensionality reduction has a minimal effect on the resulting prediction performance, but clustering needs 20 times less computational time and only 0.1% of the original information. Finally, in Paper V, we look at similarities of representative day clusters recognized by speed and flows. Furthermore, the interchangeability of speed day-type centroids for flow when predicting speeds has proven to be robust, which is not a case for predicting flows by speed day-type centroids and observations.Trängsel i storstäderna leder till extra restid, buller, luftföroreningar, koldioxidutsläpp med mera. Transporter är en av de främsta erkända bidragsgivarna till global uppvärmning och klimatförändringar, som får allt större uppmärksamhet från myndigheter och samhällen runt om i världen. Bättre utnyttjande av befintliga resurser genom intelligenta transportsystem (ITS) och digital teknik identifieras av Europeiska kommissionen som teknik med en enorm potential att minska ovanstående negativa effekter kopplade till stora trafikvolymer i stadsområden. Huvudfokus i detta arbete ligger påkortsiktiga trafikprognoser, som är ett viktigt verktyg inom ITS. I kombination med informationsförsörjning möjliggör de proaktiva beslut för att minska omfattningen av trafikstockningar som uppstår regelbundet eller orsakas av incidenter. Det viktigaste bidraget i detta arbete är att utveckla ett metodologiskt ramverk och bevisa dess förbättrande effekter påkortsiktiga prognoser för storskaliga transportnät. Det förväntas bidra till mer robusta och exakta prognoser av ITS i trafikledningscentraler. Trafikmönster i storskaliga nät, inklusive stadsgator, kan vara heterogena under dagen och från dag till dag. I detta arbete undersöks rumslig och temporal klustring av heterogena datamängder till mindre, mer homogena datamängder. Detta förväntas ge mer robusta, exakta, skalbara och kostnadseffektiva prognosmodeller. Avhandlingen är en samling av fem artiklar som bidrar till att förbättra kortsiktiga trafikprognoser i detta sammanhang. Klustring påvisas öka prediktionsprestandan i artiklar II, III, IV och V. I artikel II beaktas nätverksuppdelning och i de tre sista dokumenten klusterbildning. De prediktionsmodeller som används i de inkluderade artiklarna är naiva historiska medelvärdesprediktionsmodeller och mer avancerade parametriska prediktionsmodeller, t.ex. probabilistisk principalkomponentanalys (PPCA) och exponentiell utjämning. I artikel I beaktas och utnyttjas probfordonsdata (FCD) som en kostnadseffektiv opportunistisk källa till hastighets- och restidsdata med omfattande nätverkstäckning. Den vedertagna metoden för att bestämma antalet kluster är att förlita sig påinterna utvärderingsindex, och dessa är mycket effektiva men isolerade från tillämpningen. I uppsats IV testas denna praxis genom att även beakta prestandan i en tillämpning för korttidsprognoser. Våra resultat visar att om man förlitar sig pådessa index kan det leda till en förlust av prediktionsprestanda påcirka 20% beroende påvilken prognosmodell som används. Dimensionalitetsminskning har en minimal effekt påden resulterande prediktionsprestandan, men klusterbildning kräver 20 gånger mindre beräkningstid och endast 0,1% av den ursprungliga informationen. Slutligen undersöker vi i artikel V likheterna mellan representativa dagskluster som bildas genom hastighet respektive flöden. Dessutom visar sig utbytbarheten av dagstypcentroider från hastigheter till flöden robust vid prediktion av hastigheter , vilket inte är fallet när det gäller prediktion av flöden
Odhad momentů při intervalovém cenzorování typu I
No full textTitle: Moments Estimation under Type I Interval Censoring Author: Matej Ďurčík Department: Faculty of Probability and Mathematical Statistics Supervisor: RNDr. Arnošt Komárek Ph.D. Abstract: In this thesis we apply the uniform deconvolution model to the interval censoring problem. We restrict ourselves only on interval censoring case 1. We show how to apply uniform deconvolution model in estimating the probability distribution characteristics in the interval censoring case 1. Moreover we derive limit distributions of the estimators of mean and variance. Then we compare these estimators to the asymptotically efficient estimators based on the nonparametric maximum likelihood estimation by simulation studies under some certain distributions of the random variables.
DEVELOPMENT OF CONTENT ON DEMAND SYSTEM ON XBMC PLATFORM
No full textV diplomski nalogi predstavljamo razvoj sistema vsebin na zahtevo na XBMC platformi. V nalogi predstavimo infrastrukturo sistema IPTV, pripadajoče storitve ter module sistema. Za sistem UMB-SmartTV smo razvili tudi nov XBMC vtičnik za pregled lastnih vsebin. Vtičnik omogoča pregled in predvajanje različnih vsebin na sistemu UMB SmartTV, ki jih lahko uporabnik dodaja v bazo sistema. Podatke o vsebinah je mogoče pregledovati glede na želene kategorije: žanr, leto nastanka, režiser, igralci, avtor itd. Poudarek pri razvoju sistema smo namenili čim hitrejšemu iskanju vsebin, kar smo dosegli tudi z razvrščanjem vsebin v kategorije.The purpose of the diploma thesis is to introduce the development of a Content-on-Demand system on the XBMC platform. The thesis deals with the architecture of the IPTV system, the services that are delivered through this system and its modules. Furthermore we have developed a new XBMC plugin for the content review for the UMB-SmartTV system. The plugin allows the user to view and play various contents that can be added to the system’s database on the UMB-SmartTV system. The user is able to review the information about the content by selecting between different categories like genre, director, actors, author etc. The main goal was to develop a system that provides fast search for specific content and the classification of the content into categories that can help us to access it
Vícečetné zarovnávání sekvencí
No full textNázev práce: Vícečetné zarovnávání sekvencí Autor: Matej Ferenc Katedra (ústav): Katedra aplikované matematiky Vedoucí bakalářské práce: RNDr. Ondřej Pangrác, Ph.D. e-mail vedoucího: [email protected] Abstrakt: V práci študujeme problém zarovnania viacerých proteínových alebo DNA sekvencií. Existuje mnoho prístupov k jeho riešeniu, pričom niektoré algoritmy sú optimalizované na rýchlosť, iné na kvalitu zarovnania. Implementujeme dve metódy - iteratívnu a progresívnu, ktoré vychádzajú z rovnakého princípu: použiť vývojové stromy, pomocou ktorých zostavíme zarovnanie. Zavedieme niekoľko metód na výpočet vzdialenosti sekvencií. Cieľom práce je porovnať jednotlivé metódy pre zarovnanie a zistiť, kedy je ich vhodné použiť a tiež nájsť parametre, pomocou ktorých dosiahneme najlepšie výsledky zarovnania. Klíčová slova: bioinformatika, zarovnanie, sekvencieTitle: Multiple Sequence Alignment Author: Matej Ferenc Department: Department of Applied Mathematics Supervisor: RNDr. Ondřej Pangrác, Ph.D. Supervisor's e-mail address: [email protected] Abstract: In this work we study multiple sequence alignment problem, for aligning protein or DNA sequences. There is a lot of ways how to solve this problem. Some of them are optimized for speed, while others are optimized for quality of the produced alignment. We implement two methods - progressive and iterative, which are based on creating a phylogeny tree and align sequences according to it. We will also provide a few distance methods. Our aim is to compare the methods and their parameters for creating best alignments and to find out, when to use which methods. Keywords: bioinformatics, alignment, sequencesDepartment of Applied MathematicsKatedra aplikované matematikyFaculty of Mathematics and PhysicsMatematicko-fyzikální fakult
Hudební efekty
No full textNázov práce: Hudobné efekty Autor: Matej Marko Katedra (ústav): Katedra aplikovanej matematiky Vedúci baklárskej práce: Mgr. Martin Bálek e-mail vedúceho: [email protected] Abstrakt: Efekty zohrávajú v procese tvorby hudby významnú úlohu. Pomáhajú doplniť a obohatiť aranžmán skladby. Cieľom práce je vyvinúť aplikáciu, ktorá umožní užívateľovi prehrať zvukový súbor a v reálnom čase na prehrávaný obsah aplikovať zvolené efekty. Každá zmena v nastavení efektov sa tak ihneď prejaví na výslednom zvuku a užívateľ ju môže okamžite posúdiť. Konkrétne nastavenie môže užívateľ aplikovať na celý vstupný súbor a tento výsledok uložiť. Súčasťou aplikácie je tiež implementácia vybraných efektov. V texte práce sa potom zameriavame na postupy, ktoré sú použité pri implementovaní týchto efektov. Kľúčové slová: efekty, hudba, zvukTitle: Music effects Author: Matej Marko Department: Department of applied mathematics Supervisor: Mgr. Martin Bálek Supervisor's e-mail address: [email protected] Abstract: The effects have crucial position in the proccess of creating music. They enrich song's arrangement and make it all more colorful. Aim of the work is to create application that allows the user to play a sound file and apply chosen effets on it's content in real-time. All changes in effects' settings are imidiatelly transformed into the hearable results. Certain settings can be applied on the whole file and the results afterwards exported to the file. Part of the application is naturally implementation of chosen effects. Main topic of the work's text are methods used to implement the effects. Keywords: effects, music, soundDepartment of Applied MathematicsKatedra aplikované matematikyFaculty of Mathematics and PhysicsMatematicko-fyzikální fakult
De ratione studii: Educational vision of Erasmus of Rotterdam
No full textErazem Rotterdamski je kot ena izmed vodilnih osebnosti renesančnega humanizma napisal več vplivnih del s področja vzgoje in izobraževanja. V njih je predstavil svoja stališča o tem, kako je treba zastaviti vzgojo in pouk, da bi bili učenci deležni karseda kakovostne izobrazbe. Ta temelji predvsem na pridobivanju spretnosti v izražanju v klasičnih jezikih ter na branju antične literature, s pomočjo katere se učencu izoblikuje kreposten značaj. V svojih delih Erazem veliko pozornosti posveča pravilni usposobljenosti učitelja, ki mora biti vsestransko razgledan in načitan, pomembna pa je tudi njegova osebnost, saj mu mora biti učenec naklonjen, da učenje poteka čim učinkoviteje. Erazem se v svojih nasvetih zgleduje predvsem po antičnih in renesančnih piscih, saj meni, da je v obdobju srednjega veka kakovost izobraževanja upadla. Pričujoče magistrsko delo vsebuje tudi prevod Erazmovega priročnika za učitelje, naslovljenega O metodi pouka, v katerem povzema svoja stališča glede različnih vidikov izobraževanja.Erasmus of Rotterdam is one of the leading figures of the Renaissance humanism. He wrote several influential works concerning upbringing and education. They contain his views about the correct forms of raising and teaching children that are necessary for obtaining optimal education. It should be based on acquiring an advanced knowledge of classical languages and on reading classical, which helps to form a child\u27s character. In his works Erasmus often focuses on the optimal qualifications of a teacher, who needs to be well-read in a wide spectre of different disciplines. He needs to have a pleasant personality, so the children will be fond of him and the learning will be as efficient as possible. In his writings on education Erasmus follows the examples of writers from antiquity and renaissance, as he considers the education of the middle ages to be of inferior quality. The present master thesis also contains a translation of Erasmus\u27 handbook for teachers, titled On the method of study, which contains author\u27s views on various aspects of education
Multiple sequence alignment
No full textTitle: Multiple Sequence Alignment Author: Matej Ferenc Department: Department of Applied Mathematics Supervisor: RNDr. Ondřej Pangrác, Ph.D. Supervisor's e-mail address: [email protected] Abstract: In this work we study multiple sequence alignment problem, for aligning protein or DNA sequences. There is a lot of ways how to solve this problem. Some of them are optimized for speed, while others are optimized for quality of the produced alignment. We implement two methods - progressive and iterative, which are based on creating a phylogeny tree and align sequences according to it. We will also provide a few distance methods. Our aim is to compare the methods and their parameters for creating best alignments and to find out, when to use which methods. Keywords: bioinformatics, alignment, sequence
Framework for geolocation-based Android applications using Linked Data
No full textTitle: Framework for geolocation-based Android applications using Linked Data Author: Bc. Matej Snoha The aim of this thesis is to design and implement a framework for geolo- cation based mobile applications using Linked Data. Introduced are Linked Data technologies in the context of mobile application develop- ment, data modeling, and geographical queries. This work follows the software development lifecycle from requirement gathering, software analysis, design of the application framework and its individual compo- nents, up to the implementation of required functionality and subsequent deployment and evaluation of the functional application framework. The resulting implementation of the framework consists of a mobile applica- tion that displays nearby places from Linked Data datasets on a map and a cloud service with a repository of required definitions. It serves to demonstrate functionality of the theoretical part of the work in real-life scenarios
- …
