005 Computerprogrammierung, Programme, Daten
Refine
Document Type
- Article in a Periodical of the TH Wildau (11)
- Conference Paper (10)
- Master's Thesis (10)
- Article (2)
- Part of a Book (2)
- Report (2)
- Monograph (1)
Year of publication
Institute
- Fachbereich Ingenieur- und Naturwissenschaften (12)
- Fachbereich Wirtschaft, Informatik, Recht (12)
- Fachbereich Ingenieurwesen / Wirtschaftsingenieurwesen (bis 8/2014) (6)
- Wildau Institute of Technology (WIT) (5)
- Hochschulrechenzentrum (4)
- Fachbereich Wirtschaft, Verwaltung und Recht (bis 8/2014) (3)
- Hochschulbibliothek (2)
- Fachbereich Betriebswirtschaft / Wirtschaftsinformatik (bis 8/2014) (1)
- Zentrale Einrichtungen (1)
Has Fulltext
- yes (38)
Keywords
- Bibliothek (3)
- Informationssicherheit (2)
- KMU (2)
- Webanwendung (2)
- awareness (2)
- distributed system (2)
- mobile information system (2)
- multimedia guide (2)
- periodic event scheduling problem (2)
- periodic timetabling (2)
This research paper addresses the critical cybersecurity issues of healthcare IT systems. Through the use of artificial intelligence, specifically natural language processing (NLP) and advanced neural networks such as long short-term memory (LSTM), the study closely analyzes vulnerabilities documented in the National Vulnerability Database (NVD). The main objective is to identify patterns in the Common Vulnerabilities and Exposures (CVE) datasets to predict Common Vulnerability Scoring System (CVSS) scores with high accuracy, which is of utmost importance in the healthcare sector where IT failures can have catastrophic consequences.
The work aims not only to predict the potential impact on healthcare but also to prioritize the vulnerabilities according to their severity and their relevance to healthcare facilities. The vulnerability descriptions, vectors, and affected software configurations will be evaluated in detail. Through this analytical effort, the study will develop a differentiated framework for the early detection and professional management of IT security threats, aiming to strengthen the cybersecurity defenses of healthcare infrastructures. By contributing to the strategic anticipation and mitigation of risks, the research seeks to improve the resilience of critical healthcare systems and ensure the continuity and integrity of patient care.
XplainCVSS: Interpretable Machine Learning for Explainable CVSS Vulnerability Severity Prediction
(2026)
Automating CVSS scoring can reduce analyst workload, but practical deployment requires explanations that are not only interpretable, but also faithful to model decisions. We present XplainCVSS, a framework that predicts all CVSS v3.1 base metrics from NVD vulnerability descriptions and evaluates explanation faithfulness quantitatively. Using 173,202 CVEs with strictly temporal splits, we benchmark ten models from TF-IDF+XGBoost to domainspecific and multi-task transformers. Fine-tuned SecureBERT achieves the best performance (macro-F1 = 0.774; reconstructed base-score MAE = 0.934), while multi-task variants offer competitive consistency at 7–10x lower training cost. We validate TreeSHAP and gradient-based transformer attributions across all eight CVSS components, showing that highlighted features are decision-relevant. A temporal drift decomposition further shows that label-distribution shift (concept drift) is the main driver of degradation (delta-F1 = 0.079), indicating the need for periodic retraining. Riskaware evaluation yields a low 2.79% under-prediction rate for Critical vulnerabilities, supporting analyst-in-the-loop triage. Overall, XplainCVSS highlights a practical trade-off between transformer accuracy and the direct interpretability of keyword-based models.
Enterprise security teams must triage thousands of newly disclosed software vulnerabilities—catalogued as Common Vulnerabilities and Exposures (CVEs)—every day, yet the eventual remedy—vendor patch, workaround, mitigation, or no fix—is unknown at disclosure time, and actionable guidance for unpatched vulnerabilities is scattered across heterogeneous advisory sources at scale. We study this problem along three dimensions. First, we frame remedy prediction as four-class classification on 68,105 CVEs, showing that a concatenation multilayer perceptron (concat-MLP) fusing SecureBERT embeddings with structured metadata achieves F₁=0.793, outperforming text-only encoders by 9.3 pp; confidence thresholding yields 97.5 % accuracy at 94.3 % coverage. Second, we present a cross-source validation of remedy labels across Red Hat, Debian, and GitHub Security Advisories (GHSA): inter-source agreement is low (ϰ=0.007−0.369) and multi-source training cannot recover this gap, indicating systematic ontological disagreement. Third, we construct the CVE-Workaround Corpus (21,794 records from four sources) and benchmark five generation systems; surface metrics diverge sharply from semantic quality, and openweight models still lag behind frontier large language models (LLMs). Together, these contributions provide a predict-validate-act blueprint for vulnerability triage as a big-data service.
Network vulnerability assessment relies heavily on Nessus Attack Scripting Language scripts for automated security testing. Traditional approaches to generating these specialized security scripts face significant challenges in maintaining syntactic correctness and semantic relevance. This paper introduces a novel predictable template learning methodology that transforms non-deterministic Common Vulnerabilities and Exposures descriptions into deterministic Nessus Attack Scripting Language code generation through systematic placeholder preprocessing. We fine-tune DeepSeek Coder 6.7B using Low-Rank Adaptation on a carefully curated dataset of 99,143 Common Vulnerabilities and Exposures-Nessus Attack Scripting Language pairs, implementing template-based preprocessing to replace unpredictable elements with consistent placeholders. Our resulting fine-tuned model, NASLGenDSC, achieves 86 % training loss reduction with a final perplexity of 1.17 compared to zero-shot approaches. Comprehensive baseline evaluations against DeepSeek Coder, CodeLlama 7B, and StarCoder2 7B show our approach achieves 30-43% lower perplexity across 300 test evaluations, validating the effectiveness of template learning for security-critical code generation. The deterministic nature of our output enables practical deployment in production vulnerability management systems.
Cyberattacks on universities and their libraries have become a great problem over the last years. To address these problems, this thesis develops a market overview of the new field of behavioral biometric-based Identity and Access Management (IAM) software. With the help of expert interviews a.o., state-of-the-art security mechanisms of the library IAM Shibboleth were elicited. Afterwards, the market review was conducted and the findings were compared to a novel “CSSO” technology. As a result, we found that library IAMs such as Shibboleth could profit from using CSSO as a behavioral biometrics solution for a variety of reasons.
Modern, high-throughput methods for the analysis of genetic information, gene and metabolic products and their interactions offer new opportunities to gain comprehensive information on life processes. The data and knowledge generated open diverse application possibilities with enormous innovation potential. To unlock that potential skills in generating but also properly annotating the data for further data integration and analysis are needed. The data need to be made computer readable and interoperable to allow integration with existing knowledge leading to actionable biological insights. To achieve this, we need common standards and standard operating procedures as well as workflows that allow the combination of data across standards. Currently, there is a lack of experts who understand the principles and possess knowledge of the principles and relevant tools. This is a major barrier hindering the implementation of FAIR (findable, accessible, interoperable and reusable) data principles and the actual reusability of data. This is mainly due to insufficient and unequal education of the scientists and other stakeholders involved in producing and handling big data in life science that is inherently varied and complex in nature, and large in volume. Due to the interdisciplinary nature of life science research, education within this field faces numerous hurdles including institutional barriers, lack of local availability of all required expertise, as well as lack of appropriate teaching material and appropriate adaptation of curricula.
In this paper, the iLibrary system is introduced, which has been integrated and tested at the library of the Technical University of Applied Sciences in Wildau. The goal of this project is to develop an interactive information system for the improvement of virtual library services within the university’s library and in this manner make users more aware of all the benefits the library provides. The system concentrates mainly on the development and implementation of new software applications for devices such as smartphones, tablets and a multi-touch screen, as well as introducing new back-end applications for the improvement of existing services and processes.
We will explain the features of this system, provide some practical experiences, and show how it can easily be adapted to any library.
At University of Applied Sciences Wildau in Germany students have developed and evaluated a mobile and multilingual on-campus-information system, called iCampus Wildau. With the help of maps, audio tracks, and 360°-pictures a user is guided across campus and provided with current information about visible and invisible services and options. Because of its modular and general approach, the guide has been used for several, quite different purposes. One very successful application case has been a multimedia guide for the university library. This has been used for introducing new library users into using the library services, providing information about the special architecture of the library building, or even as a multimedia guide for one of the many exhibitions that find their place in the library. In this paper, we will explain the features of this application, provide some practical experiences, and show how it can easily be adapted to any library (or campus).
Sowohl das Bundesverkehrsministerium als auch das Verkehrsministerium Brandenburgs haben sich das Ziel gesetzt, den Anteil des Radverkehrs von 2017 bis 2030 zu verdoppeln. Ein wichtiger Faktor bei der Wahl des Fahrrads als Verkehrsmittel ist die subjektive Sicherheit. Diese wird stark beeinflusst durch Überholvorgänge von Kraftfahrzeugen. Seit 2020 ist in der Straßenverkehrs-Ordnung ein Mindestabstand von 1,50 m festgesetzt, den Kraftfahrzeuge innerorts beim Überholen von Radfahrenden einhalten müssen. In Forschungsarbeiten zum Überholverhalten zeigte sich, dass dieser Abstand häufig unterschritten wird und dass die Infrastruktur einen Einfluss auf den Überholabstand hat. Mit dem Überholabstandsmessgerät „OpenBikeSensor“ wurden in den letzten Jahren über 270.000 Überholvorgänge erfasst. Um die Daten auszuwerten zu können, müssen die Informationen zur Infrastruktur am Ort des Überholvorgangs aus einer anderen Quelle gewonnen werden. Ziel dieser Arbeit war herauszufinden, inwieweit die ebenfalls frei verfügbaren Daten aus der OpenStreetMap dafür geeignet sind. Dafür wurde einerseits eine Recherche im OpenStreetMap-Wiki durchgeführt, in welchem dokumentiert ist, wie Objekte zu erfassen sind. Andererseits wurden die für Berlin und Brandenburg verfügbaren Daten aus OpenBikeSensoren und der OpenStreetMap miteinander verknüpft und ausgewertet. Um die Güte von Auswertungsergebnissen einschätzen zu können, wurden sechs Analysen aus anderen Forschungsarbeiten nachvollzogen. Es zeigte sich, dass erreichbare Ergebnisse eine gute Vergleichbarkeit mit den Ergebnissen anderer Forschungen aufweisen, wenn eine hohe Anzahl von Überholvorgängen und eine detaillierte Beschreibung der untersuchten Infrastruktur vorliegen, und dass die kombinierten Daten weitreichende Analysen ermöglichen können. Die wesentliche Einschränkung liegt darin, dass für viele Straßen noch keine ausreichende Beschreibung in der OpenStreetMap verfügbar ist. Eine Tendenz zur Zunahme dieser Informationen war erkennbar.
Model based Configuration of Platforms for Managing Cross- Organizational (Business) Processes
(2022)
In this contribution interoperability is considered from the perspective of platforms, which have to manage cross-organisational business processes. A model-based approach for configuring a cloud platform for managing complex processes and their dependencies across different organisations is provided. The approach is applied on using FIWARE, which provides a framework of open source software platform components. The core concept is to extend the open source core data model of FIWARE by using the artefacts of an Enterprise model, describing the dependencies of processes, roles, object data and application interfaces. Based on a given use case the principal configuration was applied and validated.
