Refine
Has Fulltext
- yes (30)
Is part of the Bibliography
- yes (30)
Year of publication
- 2020 (30) (remove)
Document Type
- Journal article (15)
- Doctoral Thesis (9)
- Working Paper (3)
- Report (2)
- Bachelor Thesis (1)
Keywords
- Crowdsourcing (3)
- Quality of Experience (3)
- Echtzeit (2)
- Ethernet (2)
- Latenz (2)
- Optical Character Recognition (2)
- QoE (2)
- Rechnernetz (2)
- endurance (2)
- jitter (2)
- self-adaptive systems (2)
- self-aware computing (2)
- virtual reality (2)
- 3d point clouds (1)
- Alter Druck (1)
- Automatic Text Reconition (1)
- Automatische Texterkennung (ATR) (1)
- Automatisches Kalibrieren (1)
- Avatar <Informatik> (1)
- Bestärkendes Lernen (1)
- Character Networks (1)
- Character Reference Detection (1)
- Communication (1)
- Coreference (1)
- Crowdsensing (1)
- CubeSat (1)
- Datenübertragung (1)
- Deep Learning (1)
- Deep Reinforcement Learning (1)
- Design patterns (1)
- Diagnosesystem (1)
- Digital Humanities (1)
- Document Analysis (1)
- Dreidimensionale Bildverarbeitung (1)
- Echtzeit-Netzwerke (1)
- Energieversorgung (1)
- Enterprise application (1)
- Enterprise-Resource-Planning (1)
- Entscheidungsunterstützungssystem (1)
- Expected MOS (1)
- Expected QoE (1)
- FPGA (1)
- Field programmable gate array (1)
- Formal analysis (1)
- Formal verification (1)
- GNSS/INS integrated navigation (1)
- Geschäftsanwendung (1)
- Good-or-Better (GoB) (1)
- Hardware (1)
- Historical Printings (1)
- INS/LIDAR integrated navigation (1)
- Internet of Things (1)
- IoT (1)
- Künstliche Intelligenz (1)
- Latency Bound (1)
- Mensch-Maschine-Kommunikation (1)
- Modellbasierte Diagnose (1)
- Monitoring (1)
- NanoFEEP (1)
- Network Measurements (1)
- Netzwerk (1)
- Netzwerkanalyse <Soziologie> (1)
- Neume Notation (1)
- Neumennotation (1)
- Neumenschrift (1)
- Nutzerstudien (1)
- Optical Music Recognition (1)
- Optische Musikerkennung (OMR) (1)
- Optische Zeichenerkennung (1)
- Optische Zeichenerkennung (OCR) (1)
- Punktwolke (1)
- QoE estimation (1)
- QoE fundamentals (1)
- QoE-Abschätzung (1)
- QoS-QoE mapping functions (1)
- Quality of Experience (QoE) (1)
- Quality of Service (QoS) (1)
- Quotation Attribution (1)
- Raumfahrttechnik (1)
- Real-Time-Networks (1)
- Reinforcement Learning (1)
- Relation Detection (1)
- SDN (1)
- ST-elevation myocardial infarction (1)
- Satellit (1)
- Service-level Quality Index (SQI) (1)
- Software-defined networking (1)
- SpaceWire (1)
- TSN (1)
- Textanalyse (1)
- Time-Sensitive Networking (1)
- Time-Sensitive-Networking (1)
- UWE-4 (1)
- User studies (1)
- Verdachtsüberprüfung (1)
- Verifikation (1)
- Virtuelle Realität (1)
- aerospace engineering (1)
- anomaly detection (1)
- anomaly prediction (1)
- arithmetic calculations (1)
- artificial intelligence (1)
- augmented reality (1)
- autonomous (1)
- avatars (1)
- biomechanic (1)
- biomechanical engineering (1)
- biomimetics (1)
- blood coagulation factor XIII (1)
- cardiac magnetic resonance imaging (1)
- cardiorespiratory fitness (1)
- change detection (1)
- characterization (1)
- collaborative interaction (1)
- collision detection (1)
- continuous-time SLAM (1)
- crowdsourced QoE measurements (1)
- crowdsourced network measurements (1)
- crowdsourcing (1)
- cybersickness (1)
- deformation-based method (1)
- distraction (1)
- electric propulsion (1)
- exercise intensity (1)
- experimental evaluation (1)
- failure prediction (1)
- granular (1)
- healing and remodelling processes (1)
- hybrid avatar-agent systems (1)
- in-orbit experiments (1)
- intelligent transportation systems (1)
- k-d tree (1)
- latency (1)
- locomotion (1)
- machine learning (1)
- mapping (1)
- measurements (1)
- mechanical engineering (1)
- mechanics (1)
- multirotors (1)
- neural architecture (1)
- neural networks (1)
- performance prediction (1)
- personalized training (1)
- power consumption (1)
- precision training (1)
- prediction (1)
- quadcopters (1)
- quality assurance (1)
- quality evaluation (1)
- sandfish (1)
- self-adaptive (1)
- self-aware computing systems (1)
- self-managing systems (1)
- simulator sickness (1)
- slip (1)
- social artificial intelligence (1)
- software engineering (1)
- study design (1)
- survey (1)
- switching navigation (1)
- taxonomy (1)
- thrust direction (1)
- time calibration (1)
- traffic damping (1)
- unmanned aerial vehicles (1)
- user study (1)
- vehicular navigation (1)
- verification of suspected diagnoses (1)
- video QoE (1)
- virtual embodiment (1)
- virtual social interaction (1)
- wearable (1)
- Änderungserkennung (1)
Institute
- Institut für Informatik (30) (remove)
Sonstige beteiligte Institutionen
EU-Project number / Contract (GA) number
- 320377 (1)
White Paper on Crowdsourced Network and QoE Measurements – Definitions, Use Cases and Challenges
(2020)
The goal of the white paper at hand is as follows. The definitions of the terms build a framework for discussions around the hype topic ‘crowdsourcing’. This serves as a basis for differentiation and a consistent view from different perspectives on crowdsourced network measurements, with the goal to provide a commonly accepted definition in the community. The focus is on the context of mobile and fixed network operators, but also on measurements of different layers (network, application, user layer). In addition, the white paper shows the value of crowdsourcing for selected use cases, e.g., to improve QoE or regulatory issues. Finally, the major challenges and issues for researchers and practitioners are highlighted.
This white paper is the outcome of the Würzburg seminar on “Crowdsourced Network and QoE Measurements” which took place from 25-26 September 2019 in Würzburg, Germany. International experts were invited from industry and academia. They are well known in their communities, having different backgrounds in crowdsourcing, mobile networks, network measurements, network performance, Quality of Service (QoS), and Quality of Experience (QoE). The discussions in the seminar focused on how crowdsourcing will support vendors, operators, and regulators to determine the Quality of Experience in new 5G networks that enable various new applications and network architectures. As a result of the discussions, the need for a white paper manifested, with the goal of providing a scientific discussion of the terms “crowdsourced network measurements” and “crowdsourced QoE measurements”, describing relevant use cases for such crowdsourced data, and its underlying challenges. During the seminar, those main topics were identified, intensively discussed in break-out groups, and brought back into the plenum several times. The outcome of the seminar is this white paper at hand which is – to our knowledge – the first one covering the topic of crowdsourced network and QoE measurements.
The electric propulsion system NanoFEEP was integrated and tested in orbit on the UWE-4 satellite, which marks the first successful demonstration of an electric propulsion system on board a 1U CubeSat. In-orbit characterization measurements of the heating process of the propellant and the power consumption of the propulsion system at different thrust levels are presented. Furthermore, an analysis of the thrust vector direction based on its effect on the attitude of the spacecraft is described. The employed heater liquefies the propellant for a duration of 30 min per orbit and consumes 103 ± 4 mW. During this time, the respective thruster can be activated. The propulsion system including one thruster head, its corresponding heater, the neutralizer and the digital components of the power processing unit consume 8.5 ± 0.1 mW ⋅μ A\(^{−1}\) + 184 ± 8.5 mW and scales with the emitter current. The estimated thrust directions of two thruster heads are at angles of 15.7 ± 7.6∘ and 13.2 ± 5.5∘ relative to their mounting direction in the CubeSat structure. In light of the very limited power on a 1U CubeSat, the NanoFEEP propulsion system renders a very viable option. The heater of subsequent NanoFEEP thrusters was already improved, such that the system can be activated during the whole orbit period.
In the present day, unmanned aerial vehicles become seemingly more popular every year, but, without regulation of the increasing number of these vehicles, the air space could become chaotic and uncontrollable. In this work, a framework is proposed to combine self-aware computing with multirotor formations to address this problem. The self-awareness is envisioned to improve the dynamic behavior of multirotors. The formation scheme that is implemented is called platooning, which arranges vehicles in a string behind the lead vehicle and is proposed to bring order into chaotic air space. Since multirotors define a general category of unmanned aerial vehicles, the focus of this thesis are quadcopters, platforms with four rotors. A modification for the LRA-M self-awareness loop is proposed and named Platooning Awareness. The implemented framework is able to offer two flight modes that enable waypoint following and the self-awareness module to find a path through scenarios, where obstacles are present on the way, onto a goal position. The evaluation of this work shows that the proposed framework is able to use self-awareness to learn about its environment, avoid obstacles, and can successfully move a platoon of drones through multiple scenarios.
Dessert organisms like sandfish lizards (SLs) bend and generate thrust in granular mediums to scape heat and hunt for prey [1]. Further, SLs seems to have striking capabilities to swim in undulatory form keeping the same wavelength even in terrains with different volumetric densities, hence behaving as rigid bodies. This paper tries to recommend new research directions for planetary robotics, adapting principles of sand swimmers for improving robustness of surface exploration robots. First, we summarize previous efforts on bio-inspired hardware developed for granular terrains and accessing complex geological features. Later, a rigid wheel design has been proposed to imitate SLs locomotion capabilities. In order to derive the force models to predict performance of such bio-inspired mobility system, different approaches as RFT (Resistive Force Theory) and analytical terramechanics are introduced. Even in typical wheeled robots the slip and sinkage increase with time, the new design intends to imitate traversability capabilities of SLs, that seem to keep the same slip while displacing at subsurface levels.
Recent advances in Natural Language Preprocessing (NLP) allow for a fully automatic extraction of character networks for an incoming text. These networks serve as a compact and easy to grasp representation of literary fiction. They offer an aggregated view of the text, which can be used during distant reading approaches for the analysis of literary hypotheses. In their core, the networks consist of nodes, which represent literary characters, and edges, which represent relations between characters. For an automatic extraction of such a network, the first step is the detection of the references of all fictional entities that are of importance for a text. References to the fictional entities appear in the form of names, noun phrases and pronouns and prior to this work, no components capable of automatic detection of character references were available. Existing tools are only capable of detecting proper nouns, a subset of all character references. When evaluated on the task of detecting proper nouns in the domain of literary fiction, they still underperform at an F1-score of just about 50%. This thesis uses techniques from the field of semi-supervised learning, such as Distant supervision and Generalized Expectations, and improves the results of an existing tool to about 82%, when evaluated on all three categories in literary fiction, but without the need for annotated data in the target domain. However, since this quality is still not sufficient, the decision to annotate DROC, a corpus comprising 90 fragments of German novels was made. This resulted in a new general purpose annotation environment titled as ATHEN, as well as annotated data that spans about 500.000 tokens in total. Using this data, the combination of supervised algorithms and a tailored rule based algorithm, which in combination are able to exploit both - local consistencies as well as global consistencies - yield an algorithm with an F1-score of about 93%. This component is referred to as the Kallimachos tagger.
A character network can not directly display references however, instead they need to be clustered so that all references that belong to a real world or fictional entity are grouped together. This process widely known as coreference resolution is a hard problem in the focus of research for more than half a century. This work experimented with adaptations of classical feature based machine learning, with a dedicated rule based algorithm and with modern techniques of Deep Learning, but no approach can surpass 55% B-Cubed F1, when evaluated on DROC. Due to this barrier, many researchers do not use a fully-fledged coreference resolution when they extract character networks, but only focus on a more forgiving subset- the names. For novels such as Alice's Adventures in Wonderland by Lewis Caroll, this would however only result in a network in which many important characters are missing. In order to integrate important characters into the network that are not named by the author, this work makes use of automatic detection of speaker and addressees for direct speech utterances (all entities involved in a dialog are considered to be of importance). This problem is by itself not an easy task, however the most successful system analysed in this thesis is able to correctly determine the speaker to about 85% of the utterances as well as about 65% of the addressees. This speaker information can not only help to identify the most dominant characters, but also serves as a way to model the relations between entities.
During the span of this work, components have been developed to model relations between characters using speaker attribution, using co-occurrences as well as by the usage of true interactions, for which yet again a dataset was annotated using ATHEN. Furthermore, since relations between characters are usually typed, a component for the extraction of a typed relation was developed. Similar to the experiments for the character reference detection, a combination of a rule based and a Maximum Entropy classifier yielded the best overall results, with the extraction of family relations showing a score of about 80% and the quality of love relations with a score of about 50%. For family relations, a kernel for a Support Vector Machine was developed that even exceeded the scores of the combined approach but is behind on the other labels.
In addition, this work presents new ways to evaluate automatically extracted networks without the need of domain experts, instead it relies on the usage of expert summaries. It also refrains from the uses of social network analysis for the evaluation, but instead presents ranked evaluations using Precision@k and the Spearman Rank correlation coefficient for the evaluation of the nodes and edges of the network. An analysis using these metrics showed, that the central characters of a novel are contained with high probability but the quality drops rather fast if more than five entities are analyzed. The quality of the edges is mainly dominated by the quality of the coreference resolution and the correlation coefficient between gold edges and system edges therefore varies between 30 and 60%.
All developed components are aggregated alongside a large set of other preprocessing modules in the Kallimachos pipeline and can be reused without any restrictions.
The DFG project “SDN-enabled Application-aware Network Control Architectures and their Performance Assessment” (DFG SDN-App) focused in phase 1 (Jan 2017 – Dec 2019) on software defined networking (SDN). Being a fundamental paradigm shift, SDN enables a remote control of networking devices made by different vendors from a logically centralized controller. In principle, this enables a more dynamic and flexible management of network resources compared to the traditional legacy networks. Phase 1 focused on multimedia applications and their users’ Quality of Experience (QoE).
This documents reports the achievements of the first phase (Jan 2017 – Dec 2019), which is jointly carried out by the Technical University of Munich, Technical University of Berlin, and University of Würzburg. The project started at the institutions in Munich and Würzburg in January 2017 and lasted until December 2019.
In Phase 1, the project targeted the development of fundamental control mechanisms for network-aware application control and application-aware network control in Software Defined Networks (SDN) so to enhance the user perceived quality (QoE). The idea is to leverage the QoE from multiple applications as control input parameter for application-and network control mechanisms. These mechanisms are implemented by an Application Control Plane (ACP) and a Network Control Plane (NCP). In order to obtain a global view of the current system state, applications and network parameters are monitored and communicated to the respective control plane interface. Network and application information and their demands are exchanged between the control planes so to derive appropriate control actions. To this end, a methodology is developed to assess the application performance and in particular the QoE. This requires an appropriate QoE modeling of the applications considered in the project as well as metrics like QoE fairness to be utilized within QoE management.
In summary, the application-network interaction can improve the QoE for multi-application scenarios. This is ensured by utilizing information from the application layer, which are mapped by appropriate QoS-QoE models to QoE within a network control plane. On the other hand, network information is monitored and communicated to the application control plane. Network and application information and their demands are exchanged between the control planes so to derive appropriate control actions.
Bridge-local latency computation is often regarded with caution, as historic efforts with the Credit-Based Shaper (CBS) showed that CBS requires network wide information for tight bounds. Recently, new shaping mechanisms and timed gates were applied to achieve such guarantees nonetheless, but they require support for these new mechanisms in the forwarding devices.
This document presents a per-hop latency bound for individual streams in a class-based network that applies the IEEE 802.1Q strict priority transmission selection algorithm. It is based on self-pacing talkers and uses the accumulated latency fields during the reservation process to provide upper bounds with bridge-local information. The presented delay bound is proven mathematically and then evaluated with respect to its accuracy. It indicates the required information that must be provided for admission control, e.g., implemented by a resource reservation protocol such as IEEE 802.1Qdd. Further, it hints at potential improvements regarding new mechanisms and higher accuracy given more information.
The rating of perceived exertion (RPE) is a subjective load marker and may assist in individualizing training prescription, particularly by adjusting running intensity. Unfortunately, RPE has shortcomings (e.g., underreporting) and cannot be monitored continuously and automatically throughout a training sessions. In this pilot study, we aimed to predict two classes of RPE (≤15 “Somewhat hard to hard” on Borg’s 6–20 scale vs. RPE >15 in runners by analyzing data recorded by a commercially-available smartwatch with machine learning algorithms. Twelve trained and untrained runners performed long-continuous runs at a constant self-selected pace to volitional exhaustion. Untrained runners reported their RPE each kilometer, whereas trained runners reported every five kilometers. The kinetics of heart rate, step cadence, and running velocity were recorded continuously ( 1 Hz ) with a commercially-available smartwatch (Polar V800). We trained different machine learning algorithms to estimate the two classes of RPE based on the time series sensor data derived from the smartwatch. Predictions were analyzed in different settings: accuracy overall and per runner type; i.e., accuracy for trained and untrained runners independently. We achieved top accuracies of 84.8 % for the whole dataset, 81.8 % for the trained runners, and 86.1 % for the untrained runners. We predict two classes of RPE with high accuracy using machine learning and smartwatch data. This approach might aid in individualizing training prescriptions.
Von technischen Systemen wird in der heutigen Zeit erwartet, dass diese stets fehlerfrei funktionieren, um einen reibungslosen Ablauf des Alltags zu gewährleisten. Technische Systeme jedoch können Defekte aufweisen, die deren Funktionsweise einschränken oder zu deren Totalausfall führen können. Grundsätzlich zeigen sich Defekte durch eine Veränderung im Verhalten von einzelnen Komponenten. Diese Abweichungen vom Nominalverhalten nehmen dabei an Intensität zu, je näher die entsprechende Komponente an einem Totalausfall ist. Aus diesem Grund sollte das Fehlverhalten von Komponenten rechtzeitig erkannt werden, um permanenten Schaden zu verhindern. Von besonderer Bedeutung ist dies für die Luft- und Raumfahrt. Bei einem Satelliten kann keine Wartung seiner Komponenten durchgeführt werden, wenn er sich bereits im Orbit befindet. Der Defekt einer einzelnen Komponente, wie der Batterie der Energieversorgung, kann hierbei den Verlust der gesamten Mission bedeuten. Grundsätzlich lässt sich Fehlererkennung manuell durchführen, wie es im Satellitenbetrieb oft üblich ist. Hierfür muss ein menschlicher Experte, ein sogenannter Operator, das System überwachen. Diese Form der Überwachung ist allerdings stark von der Zeit, Verfügbarkeit und Expertise des Operators, der die Überwachung durchführt, abhängig. Ein anderer Ansatz ist die Verwendung eines dedizierten Diagnosesystems. Dieses kann das technische System permanent überwachen und selbstständig Diagnosen berechnen. Die Diagnosen können dann durch einen Experten eingesehen werden, der auf ihrer Basis Aktionen durchführen kann. Das in dieser Arbeit vorgestellte modellbasierte Diagnosesystem verwendet ein quantitatives Modell eines technischen Systems, das dessen Nominalverhalten beschreibt. Das beobachtete Verhalten des technischen Systems, gegeben durch Messwerte, wird mit seinem erwarteten Verhalten, gegeben durch simulierte Werte des Modells, verglichen und Diskrepanzen bestimmt. Jede Diskrepanz ist dabei ein Symptom. Diagnosen werden dadurch berechnet, dass zunächst zu jedem Symptom eine sogenannte Konfliktmenge berechnet wird. Dies ist eine Menge von Komponenten, sodass der Defekt einer dieser Komponenten das entsprechende Symptom erklären könnte. Mithilfe dieser Konfliktmengen werden sogenannte Treffermengen berechnet. Eine Treffermenge ist eine Menge von Komponenten, sodass der gleichzeitige Defekt aller Komponenten dieser Menge alle beobachteten Symptome erklären könnte. Jede minimale Treffermenge entspricht dabei einer Diagnose. Zur Berechnung dieser Mengen nutzt das Diagnosesystem ein Verfahren, bei dem zunächst abhängige Komponenten bestimmt werden und diese von symptombehafteten Komponenten belastet und von korrekt funktionierenden Komponenten entlastet werden. Für die einzelnen Komponenten werden Bewertungen auf Basis dieser Be- und Entlastungen berechnet und mit ihnen Diagnosen gestellt. Da das Diagnosesystem auf ausreichend genaue Modelle angewiesen ist und die manuelle Kalibrierung dieser Modelle mit erheblichem Aufwand verbunden ist, wurde ein Verfahren zur automatischen Kalibrierung entwickelt. Dieses verwendet einen Zyklischen Genetischen Algorithmus, um mithilfe von aufgezeichneten Werten der realen Komponenten Modellparameter zu bestimmen, sodass die Modelle die aufgezeichneten Daten möglichst gut reproduzieren können. Zur Evaluation der automatischen Kalibrierung wurden ein Testaufbau und verschiedene dynamische und manuell schwierig zu kalibrierende Komponenten des Qualifikationsmodells eines realen Nanosatelliten, dem SONATE-Nanosatelliten modelliert und kalibriert. Der Testaufbau bestand dabei aus einem Batteriepack, einem Laderegler, einem Tiefentladeschutz, einem Entladeregler, einem Stepper Motor HAT und einem Motor. Er wurde zusätzlich zur automatischen Kalibrierung unabhängig manuell kalibriert. Die automatisch kalibrierten Satellitenkomponenten waren ein Reaktionsrad, ein Entladeregler, Magnetspulen, bestehend aus einer Ferritkernspule und zwei Luftspulen, eine Abschlussleiterplatine und eine Batterie. Zur Evaluation des Diagnosesystems wurde die Energieversorgung des Qualifikationsmodells des SONATE-Nanosatelliten modelliert. Für die Batterien, die Entladeregler, die Magnetspulen und die Reaktionsräder wurden die vorher automatisch kalibrierten Modelle genutzt. Für das Modell der Energieversorgung wurden Fehler simuliert und diese diagnostiziert. Die Ergebnisse der Evaluation der automatischen Kalibrierung waren, dass die automatische Kalibrierung eine mit der manuellen Kalibrierung vergleichbare Genauigkeit für den Testaufbau lieferte und diese sogar leicht übertraf und dass die automatisch kalibrierten Satellitenkomponenten eine durchweg hohe Genauigkeit aufwiesen und damit für den Einsatz im Diagnosesystem geeignet waren. Die Ergebnisse der Evaluation des Diagnosesystems waren, dass die simulierten Fehler zuverlässig gefunden wurden und dass das Diagnosesystem in der Lage war die plausiblen Ursachen dieser Fehler zu diagnostizieren.
In recent years, great progress has been made in the area of Artificial Intelligence (AI) due to the possibilities of Deep Learning which steadily yielded new state-of-the-art results especially in many image recognition tasks.
Currently, in some areas, human performance is achieved or already exceeded.
This great development already had an impact on the area of Optical Music Recognition (OMR) as several novel methods relying on Deep Learning succeeded in specific tasks.
Musicologists are interested in large-scale musical analysis and in publishing digital transcriptions in a collection enabling to develop tools for searching and data retrieving.
The application of OMR promises to simplify and thus speed-up the transcription process by either providing fully-automatic or semi-automatic approaches.
This thesis focuses on the automatic transcription of Medieval music with a focus on square notation which poses a challenging task due to complex layouts, highly varying handwritten notations, and degradation.
However, since handwritten music notations are quite complex to read, even for an experienced musicologist, it is to be expected that even with new techniques of OMR manual corrections are required to obtain the transcriptions.
This thesis presents several new approaches and open source software solutions for layout analysis and Automatic Text Recognition (ATR) for early documents and for OMR of Medieval manuscripts providing state-of-the-art technology.
Fully Convolutional Networks (FCN) are applied for the segmentation of historical manuscripts and early printed books, to detect staff lines, and to recognize neume notations.
The ATR engine Calamari is presented which allows for ATR of early prints and also the recognition of lyrics.
Configurable CNN/LSTM-network architectures which are trained with the segmentation-free CTC-loss are applied to the sequential recognition of text but also monophonic music.
Finally, a syllable-to-neume assignment algorithm is presented which represents the final step to obtain a complete transcription of the music.
The evaluations show that the performances of any algorithm is highly depending on the material at hand and the number of training instances.
The presented staff line detection correctly identifies staff lines and staves with an $F_1$-score of above $99.5\%$.
The symbol recognition yields a diplomatic Symbol Accuracy Rate (dSAR) of above $90\%$ by counting the number of correct predictions in the symbols sequence normalized by its length.
The ATR of lyrics achieved a Character Error Rate (CAR) (equivalently the number of correct predictions normalized by the sentence length) of above $93\%$ trained on 771 lyric lines of Medieval manuscripts and of 99.89\% when training on around 3.5 million lines of contemporary printed fonts.
The assignment of syllables and their corresponding neumes reached $F_1$-scores of up to $99.2\%$.
A direct comparison to previously published performances is difficult due to different materials and metrics.
However, estimations show that the reported values of this thesis exceed the state-of-the-art in the area of square notation.
A further goal of this thesis is to enable musicologists without technical background to apply the developed algorithms in a complete workflow by providing a user-friendly and comfortable Graphical User Interface (GUI) encapsulating the technical details.
For this purpose, this thesis presents the web-application OMMR4all.
Its fully-functional workflow includes the proposed state-of-the-art machine-learning algorithms and optionally allows for a manual intervention at any stage to correct the output preventing error propagation.
To simplify the manual (post-) correction, OMMR4all provides an overlay-editor that superimposes the annotations with a scan of the original manuscripts so that errors can easily be spotted.
The workflow is designed to be iteratively improvable by training better models as soon as new Ground Truth (GT) is available.