Refine
Has Fulltext
- yes (33) (remove)
Year of publication
- 2022 (33) (remove)
Document Type
- Doctoral Thesis (14)
- Monograph/Edited Volume (12)
- Postprint (6)
- Report (1)
Is part of the Bibliography
- yes (33)
Keywords
- machine learning (3)
- Cloud Computing (2)
- MOOC (2)
- digital enlightenment (2)
- digital learning platform (2)
- digitale Aufklärung (2)
- digitale Lernplattform (2)
- lebenslanges Lernen (2)
- lifelong learning (2)
- maschinelles Lernen (2)
Institute
- Hasso-Plattner-Institut für Digital Engineering GmbH (33) (remove)
Data stream processing systems (DSPSs) are a key enabler to integrate continuously generated data, such as sensor measurements, into enterprise applications. DSPSs allow to steadily analyze information from data streams, e.g., to monitor manufacturing processes and enable fast reactions to anomalous behavior. Moreover, DSPSs continuously filter, sample, and aggregate incoming streams of data, which reduces the data size, and thus data storage costs.
The growing volumes of generated data have increased the demand for high-performance DSPSs, leading to a higher interest in these systems and to the development of new DSPSs. While having more DSPSs is favorable for users as it allows choosing the system that satisfies their requirements the most, it also introduces the challenge of identifying the most suitable DSPS regarding current needs as well as future demands. Having a solution to this challenge is important because replacements of DSPSs require the costly re-writing of applications if no abstraction layer is used for application development. However, quantifying performance differences between DSPSs is a difficult task. Existing benchmarks fail to integrate all core functionalities of DSPSs and lack tool support, which hinders objective result comparisons. Moreover, no current benchmark covers the combination of streaming data with existing structured business data, which is particularly relevant for companies.
This thesis proposes a performance benchmark for enterprise stream processing called ESPBench. With enterprise stream processing, we refer to the combination of streaming and structured business data. Our benchmark design represents real-world scenarios and allows for an objective result comparison as well as scaling of data. The defined benchmark query set covers all core functionalities of DSPSs. The benchmark toolkit automates the entire benchmark process and provides important features, such as query result validation and a configurable data ingestion rate.
To validate ESPBench and to ease the use of the benchmark, we propose an example implementation of the ESPBench queries leveraging the Apache Beam software development kit (SDK). The Apache Beam SDK is an abstraction layer designed for developing stream processing applications that is applied in academia as well as enterprise contexts. It allows to run the defined applications on any of the supported DSPSs. The performance impact of Apache Beam is studied in this dissertation as well. The results show that there is a significant influence that differs among DSPSs and stream processing applications. For validating ESPBench, we use the example implementation of the ESPBench queries developed using the Apache Beam SDK. We benchmark the implemented queries executed on three modern DSPSs: Apache Flink, Apache Spark Streaming, and Hazelcast Jet. The results of the study prove the functioning of ESPBench and its toolkit. ESPBench is capable of quantifying performance characteristics of DSPSs and of unveiling differences among systems.
The benchmark proposed in this thesis covers all requirements to be applied in enterprise stream processing settings, and thus represents an improvement over the current state-of-the-art.
Privacy regulations and the physical distribution of heterogeneous data are often primary concerns for the development of deep learning models in a medical context. This paper evaluates the feasibility of differentially private federated learning for chest X-ray classification as a defense against data privacy attacks. To the best of our knowledge, we are the first to directly compare the impact of differentially private training on two different neural network architectures, DenseNet121 and ResNet50. Extending the federated learning environments previously analyzed in terms of privacy, we simulated a heterogeneous and imbalanced federated setting by distributing images from the public CheXpert and Mendeley chest X-ray datasets unevenly among 36 clients. Both non-private baseline models achieved an area under the receiver operating characteristic curve (AUC) of 0.940.94 on the binary classification task of detecting the presence of a medical finding. We demonstrate that both model architectures are vulnerable to privacy violation by applying image reconstruction attacks to local model updates from individual clients. The attack was particularly successful during later training stages. To mitigate the risk of a privacy breach, we integrated Rényi differential privacy with a Gaussian noise mechanism into local model training. We evaluate model performance and attack vulnerability for privacy budgets ε∈{1,3,6,10}�∈{1,3,6,10}. The DenseNet121 achieved the best utility-privacy trade-off with an AUC of 0.940.94 for ε=6�=6. Model performance deteriorated slightly for individual clients compared to the non-private baseline. The ResNet50 only reached an AUC of 0.760.76 in the same privacy setting. Its performance was inferior to that of the DenseNet121 for all considered privacy constraints, suggesting that the DenseNet121 architecture is more robust to differentially private training.
These days design thinking is no longer a “new approach”. Among practitioners, as well as academics, interest in the topic has gathered pace over the last two decades. However, opinions are divided over the longevity of the phenomenon: whether design thinking is merely “old wine in new bottles,” a passing trend, or still evolving as it is being spread to an increasing number of organizations and industries. Despite its growing relevance and the diffusion of design thinking, knowledge on the actual status quo in organizations remains scarce. With a new study, the research team of Prof. Uebernickel and Stefanie Gerken investigates temporal developments and changes in design thinking practices in organizations over the past six years comparing the results of the 2015 “Parts without a whole” study with current practices and future developments. Companies of all sizes and from different parts of the world participated in the survey. The findings from qualitative interviews with experts, i.e., people who have years of knowledge with design thinking, were cross-checked with the results from an exploratory analysis of the survey data. This analysis uncovers significant variances and similarities in how design thinking is interpreted and applied in businesses.
Digitale Medien sind aus unserem Alltag kaum noch wegzudenken. Einer der zentralsten Bereiche für unsere Gesellschaft, die schulische Bildung, darf hier nicht hintanstehen. Wann immer der Einsatz digital unterstützter Tools pädagogisch sinnvoll ist, muss dieser in einem sicheren Rahmen ermöglicht werden können. Die HPI Schul-Cloud ist dieser Vision gefolgt, die vom Nationalen IT-Gipfel 2016 angestoßen wurde und dem Bericht vorangestellt ist – gefolgt. Sie hat sich in den vergangenen fünf Jahren vom Pilotprojekt zur unverzichtbaren IT-Infrastruktur für zahlreiche Schulen entwickelt. Während der Corona-Pandemie hat sie für viele Tausend Schulen wichtige Unterstützung bei der Umsetzung ihres Bildungsauftrags geboten. Das Ziel, eine zukunftssichere und datenschutzkonforme Infrastruktur zur digitalen Unterstützung des Unterrichts zur Verfügung zu stellen, hat sie damit mehr als erreicht. Aktuell greifen rund 1,4 Millionen Lehrkräfte und Schülerinnen und Schüler bundesweit und an den deutschen Auslandsschulen auf die HPI Schul-Cloud zu.
It is estimated that data scientists spend up to 80% of the time exploring, cleaning, and transforming their data. A major reason for that expenditure is the lack of knowledge about the used data, which are often from different sources and have heterogeneous structures. As a means to describe various properties of data, metadata can help data scientists understand and prepare their data, saving time for innovative and valuable data analytics. However, metadata do not always exist: some data file formats are not capable of storing them; metadata were deleted for privacy concerns; legacy data may have been produced by systems that were not designed to store and handle meta- data. As data are being produced at an unprecedentedly fast pace and stored in diverse formats, manually creating metadata is not only impractical but also error-prone, demanding automatic approaches for metadata detection.
In this thesis, we are focused on detecting metadata in CSV files – a type of plain-text file that, similar to spreadsheets, may contain different types of content at arbitrary positions. We propose a taxonomy of metadata in CSV files and specifically address the discovery of three different metadata: line and cell type, aggregations, and primary keys and foreign keys.
Data are organized in an ad-hoc manner in CSV files, and do not follow a fixed structure, which is assumed by common data processing tools. Detecting the structure of such files is a prerequisite of extracting information from them, which can be addressed by detecting the semantic type, such as header, data, derived, or footnote, of each line or each cell. We propose the supervised- learning approach Strudel to detect the type of lines and cells. CSV files may also include aggregations. An aggregation represents the arithmetic relationship between a numeric cell and a set of other numeric cells. Our proposed AggreCol algorithm is capable of detecting aggregations of five arithmetic functions in CSV files. Note that stylistic features, such as font style and cell background color, do not exist in CSV files. Our proposed algorithms address the respective problems by using only content, contextual, and computational features.
Storing a relational table is also a common usage of CSV files. Primary keys and foreign keys are important metadata for relational databases, which are usually not present for database instances dumped as plain-text files. We propose the HoPF algorithm to holistically detect both constraints in relational databases. Our approach is capable of distinguishing true primary and foreign keys from a great amount of spurious unique column combinations and inclusion dependencies, which can be detected by state-of-the-art data profiling algorithms.
Duplicate detection describes the process of finding multiple representations of the same real-world entity in the absence of a unique identifier, and has many application areas, such as customer relationship management, genealogy and social sciences, or online shopping. Due to the increasing amount of data in recent years, the problem has become even more challenging on the one hand, but has led to a renaissance in duplicate detection research on the other hand.
This thesis examines the effects and opportunities of transitive relationships on the duplicate detection process. Transitivity implies that if record pairs ⟨ri,rj⟩ and ⟨rj,rk⟩ are classified as duplicates, then also record pair ⟨ri,rk⟩ has to be a duplicate. However, this reasoning might contradict with the pairwise classification, which is usually based on the similarity of objects. An essential property of similarity, in contrast to equivalence, is that similarity is not necessarily transitive.
First, we experimentally evaluate the effect of an increasing data volume on the threshold selection to classify whether a record pair is a duplicate or non-duplicate. Our experiments show that independently of the pair selection algorithm and the used similarity measure, selecting a suitable threshold becomes more difficult with an increasing number of records due to an increased probability of adding a false duplicate to an existing cluster. Thus, the best threshold changes with the dataset size, and a good threshold for a small (possibly sampled) dataset is not necessarily a good threshold for a larger (possibly complete) dataset. As data grows over time, earlier selected thresholds are no longer a suitable choice, and the problem becomes worse for datasets with larger clusters.
Second, we present with the Duplicate Count Strategy (DCS) and its enhancement DCS++ two alternatives to the standard Sorted Neighborhood Method (SNM) for the selection of candidate record pairs. DCS adapts SNMs window size based on the number of detected duplicates and DCS++ uses transitive dependencies to save complex comparisons for finding duplicates in larger clusters. We prove that with a proper (domain- and data-independent!) threshold, DCS++ is more efficient than SNM without loss of effectiveness.
Third, we tackle the problem of contradicting pairwise classifications. Usually, the transitive closure is used for pairwise classifications to obtain a transitively closed result set. However, the transitive closure disregards negative classifications. We present three new and several existing clustering algorithms and experimentally evaluate them on various datasets and under various algorithm configurations. The results show that the commonly used transitive closure is inferior to most other clustering algorithms, especially for the precision of results. In scenarios with larger clusters, our proposed EMCC algorithm is, together with Markov Clustering, the best performing clustering approach for duplicate detection, although its runtime is longer than Markov Clustering due to the subexponential time complexity. EMCC especially outperforms Markov Clustering regarding the precision of the results and additionally has the advantage that it can also be used in scenarios where edge weights are not available.
Data profiling is the extraction of metadata from relational databases. An important class of metadata are multi-column dependencies. They come associated with two computational tasks. The detection problem is to decide whether a dependency of a given type and size holds in a database. The discovery problem instead asks to enumerate all valid dependencies of that type. We investigate the two problems for three types of dependencies: unique column combinations (UCCs), functional dependencies (FDs), and inclusion dependencies (INDs).
We first treat the parameterized complexity of the detection variants. We prove that the detection of UCCs and FDs, respectively, is W[2]-complete when parameterized by the size of the dependency. The detection of INDs is shown to be one of the first natural W[3]-complete problems. We further settle the enumeration complexity of the three discovery problems by presenting parsimonious equivalences with well-known enumeration problems. Namely, the discovery of UCCs is equivalent to the famous transversal hypergraph problem of enumerating the hitting sets of a hypergraph. The discovery of FDs is equivalent to the simultaneous enumeration of the hitting sets of multiple input hypergraphs. Finally, the discovery of INDs is shown to be equivalent to enumerating the satisfying assignments of antimonotone, 3-normalized Boolean formulas.
In the remainder of the thesis, we design and analyze discovery algorithms for unique column combinations. Since this is as hard as the general transversal hypergraph problem, it is an open question whether the UCCs of a database can be computed in output-polynomial time in the worst case. For the analysis, we therefore focus on instances that are structurally close to databases in practice, most notably, inputs that have small solutions. The equivalence between UCCs and hitting sets transfers the computational hardness, but also allows us to apply ideas from hypergraph theory to data profiling. We devise an discovery algorithm that runs in polynomial space on arbitrary inputs and achieves polynomial delay whenever the maximum size of any minimal UCC is bounded. Central to our approach is the extension problem for minimal hitting sets, that is, to decide for
a set of vertices whether they are contained in any minimal solution. We prove that this is yet another problem that is complete for the complexity class W[3], when parameterized by the size of the set that is to be extended. We also give several conditional lower bounds under popular hardness conjectures such as the Strong Exponential Time Hypothesis (SETH). The lower bounds suggest that the running time of our algorithm for the extension problem is close to optimal.
We further conduct an empirical analysis of our discovery algorithm on real-world databases to confirm that the hitting set perspective on data profiling has merits also in practice. We show that the resulting enumeration times undercut their theoretical worst-case bounds on practical data, and that the memory consumption of our method is much smaller than that of previous solutions. During the analysis we make two observations about the connection between databases and their corresponding hypergraphs. On the one hand, the hypergraph representations containing all relevant information are usually significantly smaller than the original inputs. On the other hand, obtaining those hypergraphs is the actual bottleneck of any practical application. The latter often takes much longer than enumerating the solutions, which is in stark contrast to the fact that the preprocessing is guaranteed to be polynomial while the enumeration may take exponential time.
To make the first observation rigorous, we introduce a maximum-entropy model for non-uniform random hypergraphs and prove that their expected number of minimal hyperedges undergoes a phase transition with respect to the total number of edges. The result also explains why larger databases may have smaller hypergraphs. Motivated by the second observation, we present a new kind of UCC discovery algorithm called Hitting Set Enumeration with Partial Information and Validation (HPIValid). It utilizes the fast enumeration times in practice in order to speed up the computation of the corresponding hypergraph. This way, we sidestep the bottleneck while maintaining the advantages of the hitting set perspective. An exhaustive empirical evaluation shows that HPIValid outperforms the current state of the art in UCC discovery. It is capable of processing databases that were previously out of reach for data profiling.
Polyglot programming allows developers to use multiple programming languages within the same software project. While it is common to use more than one language in certain programming domains, developers also apply polyglot programming for other purposes such as to re-use software written in other languages. Although established approaches to polyglot programming come with significant limitations, for example, in terms of performance and tool support, developers still use them to be able to combine languages.
Polyglot virtual machines (VMs) such as GraalVM provide a new level of polyglot programming, allowing languages to directly interact with each other. This reduces the amount of glue code needed to combine languages, results in better performance, and enables tools such as debuggers to work across languages. However, only a little research has focused on novel tools that are designed to support developers in building software with polyglot VMs. One reason is that tool-building is often an expensive activity, another one is that polyglot VMs are still a moving target as their use cases and requirements are not yet well understood.
In this thesis, we present an approach that builds on existing self-sustaining programming systems such as Squeak/Smalltalk to enable exploratory programming, a practice for exploring and gathering software requirements, and re-use their extensive tool-building capabilities in the context of polyglot VMs. Based on TruffleSqueak, our implementation for the GraalVM, we further present five case studies that demonstrate how our approach helps tool developers to design and build tools for polyglot programming. We further show that TruffleSqueak can also be used by application developers to build and evolve polyglot applications at run-time and by language and runtime developers to understand the dynamic behavior of GraalVM languages and internals. Since our platform allows all these developers to apply polyglot programming, it can further help to better understand the advantages, use cases, requirements, and challenges of polyglot VMs. Moreover, we demonstrate that our approach can also be applied to other polyglot VMs and that insights gained through it are transferable to other programming systems.
We conclude that our research on tools for polyglot programming is an important step toward making polyglot VMs more approachable for developers in practice. With good tool support, we believe polyglot VMs can make it much more common for developers to take advantage of multiple languages and their ecosystems when building software.
Language developers who design domain-specific languages or new language features need a way to make fast changes to language definitions. Those fast changes require immediate feedback. Also, it should be possible to parse the developed languages quickly to handle extensive sets of code.
Parsing expression grammars provides an easy to understand method for language definitions. Packrat parsing is a method to parse grammars of this kind, but this method is unable to handle left-recursion properly. Existing solutions either partially rewrite left-recursive rules and partly forbid them, or use complex extensions to packrat parsing that are hard to understand and cost-intensive. We investigated methods to make parsing as fast as possible, using easy to follow algorithms while not losing the ability to make fast changes to grammars.
We focused our efforts on two approaches.
One is to start from an existing technique for limited left-recursion rewriting and enhance it to work for general left-recursive grammars. The second approach is to design a grammar compilation process to find left-recursion before parsing, and in this way, reduce computational costs wherever possible and generate ready to use parser classes.
Rewriting parsing expression grammars is a task that, if done in a general way, unveils a large number of cases such that any rewriting algorithm surpasses the complexity of other left-recursive parsing algorithms. Lookahead operators introduce this complexity. However, most languages have only little portions that are left-recursive and in virtually all cases, have no indirect or hidden left-recursion. This means that the distinction of left-recursive parts of grammars from components that are non-left-recursive holds great improvement potential for existing parsers.
In this report, we list all the required steps for grammar rewriting to handle left-recursion, including grammar analysis, grammar rewriting itself, and syntax tree restructuring. Also, we describe the implementation of a parsing expression grammar framework in Squeak/Smalltalk and the possible interactions with the already existing parser Ohm/S. We quantitatively benchmarked this framework directing our focus on parsing time and the ability to use it in a live programming context. Compared with Ohm, we achieved massive parsing time improvements while preserving the ability to use our parser it as a live programming tool.
The work is essential because, for one, we outlined the difficulties and complexity that come with grammar rewriting. Also, we removed the existing limitations that came with left-recursion by eliminating them before parsing.
The “HPI Future SOC Lab” is a cooperation of the Hasso Plattner Institute (HPI) and industry partners. Its mission is to enable and promote exchange and interaction between the research community and the industry partners.
The HPI Future SOC Lab provides researchers with free of charge access to a complete infrastructure of state of the art hard and software. This infrastructure includes components, which might be too expensive for an ordinary research environment, such as servers with up to 64 cores and 2 TB main memory. The offerings address researchers particularly from but not limited to the areas of computer science and business information systems. Main areas of research include cloud computing, parallelization, and In-Memory technologies.
This technical report presents results of research projects executed in 2018. Selected projects have presented their results on April 17th and November 14th 2017 at the Future SOC Lab Day events.
The availability of commercial 3D printers and matching 3D design software has allowed a wide range of users to create physical prototypes – as long as these objects are not larger than hand size. However, when attempting to create larger, "human-scale" objects, such as furniture, not only are these machines too small, but also the commonly used 3D design software is not equipped to design with forces in mind — since forces increase disproportionately with scale.
In this thesis, we present a series of end-to-end fabrication software systems that support users in creating human-scale objects. They achieve this by providing three main functions that regular "small-scale" 3D printing software does not offer: (1) subdivision of the object into small printable components combined with ready-made objects, (2) editing based on predefined elements sturdy enough for larger scale, i.e., trusses, and (3) functionality for analyzing, detecting, and fixing structural weaknesses. The presented software systems also assist the fabrication process based on either 3D printing or steel welding technology.
The presented systems focus on three levels of engineering challenges: (1) fabricating static load-bearing objects, (2) creating mechanisms that involve motion, such as kinematic installations, and finally (3) designing mechanisms with dynamic repetitive movement where power and energy play an important role.
We demonstrate and verify the versatility of our systems by building and testing human-scale prototypes, ranging from furniture pieces, pavilions, to animatronic installations and playground equipment. We have also shared our system with schools, fablabs, and fabrication enthusiasts, who have successfully created human-scale objects that can withstand with human-scale forces.
Here we present an exome-wide rare genetic variant association study for 30 blood biomarkers in 191,971 individuals in the UK Biobank. We compare gene- based association tests for separate functional variant categories to increase interpretability and identify 193 significant gene-biomarker associations. Genes associated with biomarkers were ~ 4.5-fold enriched for conferring Mendelian disorders. In addition to performing weighted gene-based variant collapsing tests, we design and apply variant-category-specific kernel-based tests that integrate quantitative functional variant effect predictions for mis- sense variants, splicing and the binding of RNA-binding proteins. For these tests, we present a computationally efficient combination of the likelihood- ratio and score tests that found 36% more associations than the score test alone while also controlling the type-1 error. Kernel-based tests identified 13% more associations than their gene-based collapsing counterparts and had advantages in the presence of gain of function missense variants. We introduce local collapsing by amino acid position for missense variants and use it to interpret associations and identify potential novel gain of function variants in PIEZO1. Our results show the benefits of investigating different functional mechanisms when performing rare-variant association tests, and demonstrate pervasive rare-variant contribution to biomarker variability.
Pictures are a medium that helps make the past tangible and preserve memories. Without context, they are not able to do so. Pictures are brought to life by their associated stories. However, the older pictures become, the fewer contemporary witnesses can tell these stories.
Especially for large, analog picture archives, knowledge and memories are spread over many people. This creates several challenges: First, the pictures must be digitized to save them from decaying and make them available to the public. Since a simple listing of all the pictures is confusing, the pictures should be structured accessibly. Second, known information that makes the stories vivid needs to be added to the pictures. Users should get the opportunity to contribute their knowledge and memories. To make this usable for all interested parties, even for older, less technophile generations, the interface should be intuitive and error-tolerant.
The resulting requirements are not covered in their entirety by any existing software solution without losing the intuitive interface or the scalability of the system.
Therefore, we have developed our digital picture archive within the scope of a bachelor project in cooperation with the Bad Harzburg-Stiftung. For the implementation of this web application, we use the UI framework React in the frontend, which communicates via a GraphQL interface with the Content Management System Strapi in the backend. The use of this system enables our project partner to create an efficient process from scanning analog pictures to presenting them to visitors in an organized and annotated way. To customize the solution for both picture delivery and information contribution for our target group, we designed prototypes and evaluated them with people from Bad Harzburg. This helped us gain valuable insights into our system’s usability and future challenges as well as requirements.
Our web application is already being used daily by our project partner. During the project, we still came up with numerous ideas for additional features to further support the exchange of knowledge.
The analysis of behavioral models such as Graph Transformation Systems (GTSs) is of central importance in model-driven engineering. However, GTSs often result in intractably large or even infinite state spaces and may be equipped with multiple or even infinitely many start graphs. To mitigate these problems, static analysis techniques based on finite symbolic representations of sets of states or paths thereof have been devised. We focus on the technique of k-induction for establishing invariants specified using graph conditions. To this end, k-induction generates symbolic paths backwards from a symbolic state representing a violation of a candidate invariant to gather information on how that violation could have been reached possibly obtaining contradictions to assumed invariants. However, GTSs where multiple agents regularly perform actions independently from each other cannot be analyzed using this technique as of now as the independence among backward steps may prevent the gathering of relevant knowledge altogether.
In this paper, we extend k-induction to GTSs with multiple agents thereby supporting a wide range of additional GTSs. As a running example, we consider an unbounded number of shuttles driving on a large-scale track topology, which adjust their velocity to speed limits to avoid derailing. As central contribution, we develop pruning techniques based on causality and independence among backward steps and verify that k-induction remains sound under this adaptation as well as terminates in cases where it did not terminate before.
Learning from failure
(2022)
Regression testing is a widespread practice in today's software industry to ensure software product quality. Developers derive a set of test cases, and execute them frequently to ensure that their change did not adversely affect existing functionality. As the software product and its test suite grow, the time to feedback during regression test sessions increases, and impedes programmer productivity: developers wait longer for tests to complete, and delays in fault detection render fault removal increasingly difficult.
Test case prioritization addresses the problem of long feedback loops by reordering test cases, such that test cases of high failure probability run first, and test case failures become actionable early in the testing process. We ask, given test execution schedules reconstructed from publicly available data, to which extent can their fault detection efficiency improved, and which technique yields the most efficient test schedules with respect to APFD?
To this end, we recover regression 6200 test sessions from the build log files of Travis CI, a popular continuous integration service, and gather 62000 accompanying changelists. We evaluate the efficiency of current test schedules, and examine the prioritization results of state-of-the-art lightweight, history-based heuristics. We propose and evaluate a novel set of prioritization algorithms, which connect software changes and test failures in a matrix-like data structure.
Our studies indicate that the optimization potential is substantial, because the existing test plans score only 30% APFD. The predictive power of past test failures proves to be outstanding: simple heuristics, such as repeating tests with failures in recent sessions, result in efficiency scores of 95% APFD. The best-performing matrix-based heuristic achieves a similar score of 92.5% APFD. In contrast to prior approaches, we argue that matrix-based techniques are useful beyond the scope of effective prioritization, and enable a number of use cases involving software maintenance.
We validate our findings from continuous integration processes by extending a continuous testing tool within development environments with means of test prioritization, and pose further research questions. We think that our findings are suited to propel adoption of (continuous) testing practices, and that programmers' toolboxes should contain test prioritization as an existential productivity tool.
Text collections, such as corpora of books, research articles, news, or business documents are an important resource for knowledge discovery. Exploring large document collections by hand is a cumbersome but necessary task to gain new insights and find relevant information. Our digitised society allows us to utilise algorithms to support the information seeking process, for example with the help of retrieval or recommender systems. However, these systems only provide selective views of the data and require some prior knowledge to issue meaningful queries and asses a system’s response. The advancements of machine learning allow us to reduce this gap and better assist the information seeking process. For example, instead of sighting countless business documents by hand, journalists and investigator scan employ natural language processing techniques, such as named entity recognition. Al-though this greatly improves the capabilities of a data exploration platform, the wealth of information is still overwhelming. An overview of the entirety of a dataset in the form of a two-dimensional map-like visualisation may help to circumvent this issue. Such overviews enable novel interaction paradigms for users, which are similar to the exploration of digital geographical maps. In particular, they can provide valuable context by indicating how apiece of information fits into the bigger picture.This thesis proposes algorithms that appropriately pre-process heterogeneous documents and compute the layout for datasets of all kinds. Traditionally, given high-dimensional semantic representations of the data, so-called dimensionality reduction algorithms are usedto compute a layout of the data on a two-dimensional canvas. In this thesis, we focus on text corpora and go beyond only projecting the inherent semantic structure itself. Therefore,we propose three dimensionality reduction approaches that incorporate additional information into the layout process: (1) a multi-objective dimensionality reduction algorithm to jointly visualise semantic information with inherent network information derived from the underlying data; (2) a comparison of initialisation strategies for different dimensionality reduction algorithms to generate a series of layouts for corpora that grow and evolve overtime; (3) and an algorithm that updates existing layouts by incorporating user feedback provided by pointwise drag-and-drop edits. This thesis also contains system prototypes to demonstrate the proposed technologies, including pre-processing and layout of the data and presentation in interactive user interfaces.
Many participants in Massive Open Online Courses are full-time employees seeking greater flexibility in their time commitment and the available learning paths. We recently addressed these requirements by splitting up our 6-week courses into three 2-week modules followed by a separate exam. Modularizing courses offers many advantages: Shorter modules are more sustainable and can be combined, reused, and incorporated into learning paths more easily. Time flexibility for learners is also improved as exams can now be offered multiple times per year, while the learning content is available independently. In this article, we answer the question of which impact this modularization has on key learning metrics, such as course completion rates, learning success, and no-show rates. Furthermore, we investigate the influence of longer breaks between modules on these metrics. According to our analysis, course modules facilitate more selective learning behaviors that encourage learners to focus on topics they are the most interested in. At the same time, participation in overarching exams across all modules seems to be less appealing compared to an integrated exam of a 6-week course. While breaks between the modules increase the distinctive appearance of individual modules, a break before the final exam further reduces initial interest in the exams. We further reveal that participation in self-paced courses as a preparation for the final exam is unlikely to attract new learners to the course offerings, even though learners' performance is comparable to instructor-paced courses. The results of our long-term study on course modularization provide a solid foundation for future research and enable educators to make informed decisions about the design of their courses.
Modeling and Formal Analysis of Meta-Ecosystems with Dynamic Structure using Graph Transformation
(2022)
The dynamics of ecosystems is of crucial importance. Various model-based approaches exist to understand and analyze their internal effects. In this paper, we model the space structure dynamics and ecological dynamics of meta-ecosystems using the formal technique of Graph Transformation (short GT). We build GT models to describe how a meta-ecosystem (modeled as a graph) can evolve over time (modeled by GT rules) and to analyze these GT models with respect to qualitative properties such as the existence of structural stabilities. As a case study, we build three GT models describing the space structure dynamics and ecological dynamics of three different savanna meta-ecosystems. The first GT model considers a savanna meta-ecosystem that is limited in space to two ecosystem patches, whereas the other two GT models consider two savanna meta-ecosystems that are unlimited in the number of ecosystem patches and only differ in one GT rule describing how the space structure of the meta-ecosystem grows. In the first two GT models, the space structure dynamics and ecological dynamics of the meta-ecosystem shows two main structural stabilities: the first one based on grassland-savanna-woodland transitions and the second one based on grassland-desert transitions. The transition between these two structural stabilities is driven by high-intensity fires affecting the tree components. In the third GT model, the GT rule for savanna regeneration induces desertification and therefore a collapse of the meta-ecosystem. We believe that GT models provide a complementary avenue to that of existing approaches to rigorously study ecological phenomena.
openHPI
(2022)
On the occasion of the 10th openHPI anniversary, this technical report provides information about the HPI MOOC platform, including its core features, technology, and architecture.
In an introduction, the platform family with all partner platforms is presented; these now amount to nine platforms, including openHPI. This section introduces openHPI as an advisor and research partner in various projects.
In the second chapter, the functionalities and common course formats of the platform are presented. The functionalities are divided into learner and admin features. The learner features section provides detailed information about performance records, courses, and the learning materials of which a course is composed: videos, texts, and quizzes. In addition, the learning materials can be enriched by adding external exercise tools that communicate with the HPI MOOC platform via the Learning Tools Interoperability (LTI) standard. Furthermore, the concept of peer assessments completed the possible learning materials.
The section then proceeds with further information on the discussion forum, a fundamental concept of MOOCs compared to traditional e-learning offers. The section is concluded with a description of the quiz recap, learning objectives, mobile applications, gameful learning, and the help desk.
The next part of this chapter deals with the admin features. The described functionality is restricted to describing the news and announcements, dashboards and statistics, reporting capabilities, research options with A/B testing, the course feed, and the TransPipe tool to support the process of creating automated or manual subtitles. The platform supports a large variety of additional features, but a detailed description of these features goes beyond the scope of this report.
The chapter then elaborates on common course formats and openHPI teaching activities at the HPI. The chapter concludes with some best practices for course design and delivery.
The third chapter provides insights into the technology and architecture behind openHPI. A special characteristic of the openHPI project is the conscious decision to operate the complete application from bare metal to platform development. Hence, the chapter starts with a section about the openHPI Cloud, including detailed information about the data center and devices, the used cloud software OpenStack and Ceph, as well as the openHPI Cloud Service provided for the HPI.
Afterward, a section on the application technology stack and development tooling describes the application infrastructure components, the used automation, the deployment pipeline, and the tools used for monitoring and alerting. The chapter is concluded with detailed information about the technology stack and concrete platform implementation details. The section describes the service-oriented Ruby on Rails application, inter-service communication, and public APIs. It also provides more information on the design system and components used in the application. The section concludes with a discussion of the original microservice architecture, where we share our insights and reasoning for migrating back to a monolithic application.
The last chapter provides a summary and an outlook on the future of digital education.