Global ETD Search

51	Analysing User Viewing Behaviour in Video Streaming Services Markou, Ioannis January 2021 (has links) The user experience offered by a video streaming service plays a fundamental role in customer satisfaction. This experience can be degraded by poor playback quality and buffering issues. These problems can be caused by a user demand that is higher than the video streaming service capacity. Resource scaling methods can increase the available resources to cover the need. However, most resource scaling systems are reactive and scale up in an automated fashion when a certain demand threshold is exceeded. During popular live streaming content, the demand can be so high that even by scaling up at the last minute, the system might still be momentarily under-provisioned, resulting in a bad user experience. The solution to this problem is proactive scaling which is event-based, using content-related information to scale up or down, according to knowledge from past events. As a result, proactive resource scaling is a key factor in providing reliable video streaming services. Users viewing habits heavily affect demand. To provide an accurate model for proactive resource scaling tools, these habits need to be modelled. This thesis provides such a forecasting model for user views that can be used by a proactive resource scaling mechanism. This model is created by applying machine learning algorithms to data from both live TV and over-the-top streaming services. To produce a model with satisfactory accuracy, numerous data attributes were considered relating to users, content and content providers. The findings of this thesis show that user viewing demand can be modelled with high accuracy, without heavily relying on user-related attributes but instead by analysing past event logs and with knowledge of the schedule of the content provider, whether it is live tv or a video streaming service. / Användarupplevelsen som erbjuds av en videostreamingtjänst spelar en grundläggande roll för kundnöjdheten. Denna upplevelse kan försämras av dålig uppspelningskvalitet och buffertproblem. Dessa problem kan orsakas av en efterfrågan från användare som är högre än videostreamingtjänstens kapacitet. Resursskalningsmetoder kan öka tillgängliga resurser för att täcka behovet. De flesta resursskalningssystem är dock reaktiva och uppskalas automatiskt när en viss behovströskel överskrids. Under populärt livestreaminginnehåll kan efterfrågan vara så hög att även genom att skala upp i sista minuten kan systemet fortfarande vara underutnyttjat tillfälligt, vilket resulterar i en dålig användarupplevelse. Lösningen på detta problem är proaktiv skalning som är händelsebaserad och använder innehållsrelaterad information för att skala upp eller ner, enligt kunskap från tidigare händelser. Som ett resultat är proaktiv resursskalning en nyckelfaktor för att tillhandahålla tillförlitliga videostreamingtjänster. Användares visningsvanor påverkar efterfrågan kraftigt. För att ge en exakt modell för proaktiva resursskalningsverktyg måste dessa vanor modelleras. Denna avhandling ger en sådan prognosmodell för användarvyer som kan användas av en proaktiv resursskalningsmekanism. Denna modell är skapad genom att använda maskininlärningsalgoritmer på data från både live-TV och streamingtjänster. För att producera en modell med tillfredsställande noggrannhet ansågs ett flertal dataattribut relaterade till användare, innehåll och innehållsleverantörer. Resultaten av den här avhandlingen visar att efterfrågan på användare kan modelleras med hög noggrannhet utan att starkt förlita sig på användarrelaterade attribut utan istället genom att analysera tidigare händelseloggar och med kunskap om innehållsleverantörens schema, vare sig det är live-tv eller tjänster för videostreaming. Computer and Information Sciences Data- och informationsvetenskap
52	How to Estimate Local Performance using Machine learning Engineering (HELP ME) : from log files to support guidance / Att estimera lokal prestanda med hjälp av maskininlärning Ekinge, Hugo January 2023 (has links) As modern systems are becoming increasingly complex, they are also becoming more and more cumbersome to diagnose and fix when things go wrong. One domain where it is very important for machinery and equipment to stay functional is in the world of medical IT, where technology is used to improve healthcare for people all over the world. This thesis aims to help with reducing downtime on critical life-saving equipment by implementing automatic analysis of system logs that without any domain experts involved can give an indication of the state that the system is in. First, a literature study was performed where three potential candidates of suitable neural network architectures was found. Next, the networks were implemented and a data pipeline for collecting and labeling training data was set up. After training the networks and testing them on a separate data set, the best performing model out of the three was based on GRU (Gated Recurrent Unit). Lastly, this model was tested on some real world system logs from two different sites, one without known issues and one with slow image import due to network issues. The results showed that it was feasible to build such a system that can give indications on external parameters such as network speed, latency and packet loss percentage using only raw system logs as input data. GRU, 1D-CNN (1-Dimensional Convolutional Neural Network) and Transformer's Encoder are the three models that were tested, and the best performing model was shown to produce correct patterns even on the real world system logs. / I takt med att moderna system ökar i komplexitet så blir de även svårare att felsöka och reparera när det uppstår problem. Ett område där det är mycket viktigt att maskiner och utrustning fungerar korrekt är inom medicinsk IT, där teknik används för att förbättra hälso- och sjukvården för människor över hela världen. Syftet med denna avhandling är att bidra till att minska tiden som kritisk livräddande utrustning inte fungerar genom att implementera automatisk analys av systemloggarna som utan hjälp av experter inom området kan ge en indikation på vilket tillstånd som systemet befinner sig i. Först genomfördes en litteraturstudie där tre lovande typer av neurala nätverk valdes ut. Sedan implementerades dessa nätverk och det sattes upp en datapipeline för insamling och märkning av träningsdata. Efter att ha tränat nätverken och testat dem på en separat datamängd så visade det sig att den bäst presterande modellen av de tre var baserad på GRU (Gated Recurrent Unit). Slutligen testades denna modell på riktiga systemloggar från två olika sjukhus, ett utan kända problem och ett där bilder importerades långsamt på grund av nätverksproblem. Resultaten visade på att det är möjligt att konstruera ett system som kan ge indikationer på externa parametrar såsom nätverkshastighet, latens och paketförlust i procent genom att enbart använda systemloggar som indata. De tre modeller som testades var GRU, 1D-CNN (1-Dimensional Convolutional Neural Network) och Transformer's Encoder. Den bäst presterande modellen visade sig kunna producera korrekta mönster även för loggdata från verkliga system. Machine learning GRU 1D-CNN Transformer log analysis parameter estimation regression performance monitoring deep learning troubleshooting support Maskininlärning GRU 1D-CNN Transformer logganalys parameteruppskattning regression prestandaövervakning djupinlärning felsökning support Computer Sciences Datavetenskap (datalogi)
53	Petrophysics and fluid mechanics of selected wells in Bredasdorp Basin South Africa Ile, Anthony January 2013 (has links) Magister Scientiae - MSc / Pressure drop within a field can be attributed to several factors. Pressure drop occurs when fractional forces cause resistance to flowing fluid through a porous medium. In this thesis, the sciences of petrophysics and rock physics were employed to develop understanding of the physical processes that occurs in reservoirs. This study focussed on the physical properties of rock and fluid in order to provide understanding of the system and the mechanism controlling its behaviour. The change in production capacity of wells E-M 1, 2, 3, 4&5 prompted further research to find out why the there will be pressure drop from the suits of wells and which well was contributing to the drop in production pressure. The E-M wells are located in the Bredasdorp Basin and the reservoirs have trapping mechanisms of stratigraphical and structural systems in a moderate to good quality turbidite channel sandstone. The basin is predominantly an elongated north-west and south-east inherited channel from the synrift sub basin and was open to relatively free marine circulation. By the southwest the basin is enclose by southern Outeniqua basin and the Indian oceans. Sedimentation into the Bredasdorp basin thus occurred predominantly down the axis of the basin with main input direction from the west. Five wells were studied E-M1, E-M2, E-M3, E-M4, and E-M5 to identify which well is susceptible to flow within this group. Setting criteria for discriminator the result generated four well as meeting the criteria except for E-M1. The failure of E-M1 reservoir well interval was in consonant with result showed by evaluation from the log, pressure and rock physics analyses for E-M1.iv Various methods in rock physics were used to identify sediments and their conditions and by applying inverse modelling (elastic impedance) the interval properties were better reflected. Also elastic impedance proved to be an economical and quicker method in describing the lithology and depositional environment in the absence of seismic trace. South Africa Block 9 Bredasdorp Basin Petrophysics Sandstone unit Shale and clay unit Shale base line Rock physics Elastic impedance Fluid mechanics Fluid substitution Well log Data Log analysis Fluid and matrix properties Well prognosis Pore/ fracture pressure gradient Cut off and summation Amplitude versus offset
54	Katalyzátory pro kladnou elektrodu kyslíko-vodíkového palivového článku / Catalysts for positive electrode of hydrogen-oxygen fuel cell Kováč, Martin January 2010 (has links) Master's thesis deals with new methods of preparing catalytic materials for positive electrode of an oxygen-hydrogen fuel cell and the influence of potassium permanganate or doping agent molar mass change on theirs attributes. Further it studies the use of proper measuring methods designed to qualify theirs attributes and the presentation of achieved results. In particular methods of linear sweep and cyclic voltammetry and the processing of data using Koutecky-Levich and Tafel plot and wave log analysis. Values of half-wave and onset potential and kinetic coefficient have been measured and calculated.
55	Om informationstekniskt bevis Ekfeldt, Jonas January 2016 (has links) Information technology evidence consists of a mix of representations of various applications of digital electronic equipment, and can be brought to the fore in all contexts that result in legal decisions. The occurrence of such evidence in legal proceedings, and other legal decision-making, is a phenomenon previously not researched within legal science in Sweden. The thesis examines some of the consequences resulting from the occurrence of information technology evidence within Swedish practical legal and judicial decision-making. The thesis has three main focal points. The first consists of a broad identification of legal problems that information technology evidence entails. The second focal point examines the legal terminology associated with information technology evidence. The third focal point consists of identifying sources of error pertaining to information technology evidence from the adjudicator’s point of view. The examination utilizes a Swedish legal viewpoint from a perspective of the public trust in courts. Conclusions include a number of legal problems in several areas, primarily in regards to the knowledge of the adjudicator, the qualification of different means of evidence and the consequences of representational evidence upon its evaluation. In order to properly evaluate information technology evidence, judges are – to a greater extent than for other types of evidence – in need of (objective) knowledge supplementary to that provided by parties and their witnesses and experts. Furthermore, the current Swedish evidence terminology has been identified as a complex of problems in and of itself. The thesis includes suggestions on certain additions to this terminology. Several sources of error have been identified as being attributable to different procedures associated with the handling of information technology evidence, in particular in relation to computer forensic investigations. There is a general need for future research focused on matters regarding both standards of proof for and evaluation of information technology evidence. In addition, a need for deeper legal scientific studies aimed at evidence theory has been identified, inter alia regarding the extent to which frequency theories are applicable in respect to information technology evidence. The need for related further discussions on future emerging areas such as negative evidence and predictive evidence are foreseen. accountability audit trail authenticity computer evidence criminal procedure data quality digital evidence digital images digital traces electronic evidence evaluation of evidence evidence law evidence terminology evidential value evidentiary facts forensic evidence handling of evidence ict-evidence informatics information quality information representation information security it-evidence law and information technology legal evidence log log analysis machine identity means of evidence personal identity reliability representation information screen dump sources of error traceability validity autenticitet bevisfakta bevismedel bevishantering bevisprövning bevisrätt bevisterminologi bevisvärdering datakvalitet digitala bevis digitala bilder digitala spår elektroniska bevis elektroniska spår felkällor filer forensiska bevis ikt informatik informationskvalitet informationsrepresentation informationsteknik informationstekniskt bevis it-bevis it-säkerhet juridiskt bevis logganalys maskinell identitet personell identitet principen om fri bevisprövning representationsinformation riktighet rättsinformatik skärmdump spårbarhet straffprocessrätt

Page generated in 0.0749 seconds