@misc{KochRiehleMueller2024, author = {Koch, Christian and Riehle, Dirk and M{\"u}ller, Katharina}, title = {Schema Inference for Tabular Data with Explainable Artificial Intelligence}, doi = {10.34646/thn/ohmdok-1560}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-15601}, pages = {2 -- 6}, year = {2024}, abstract = {Data engineering is an integral part of the data science process. It comprises tasks such as data ingestion, data transformation, and data quality assurance. In order to fulfill these tasks, schema inference is an important capability. Its goal is to detect the structure of a dataset and to derive metadata on hierarchies, data types, etc. Artificial intelligence (AI) has the potential to automate schema inference and thus increase the efficiency of the data science process. However, as government institutions are subject to special regulations, explainability of AI models can be a mandatory requirement. Goal of this research protocol is to plan a systematic review of literature on schema inference for tabular data with explainable AI (XAI). This third version was derived from two earlier review protocols.}, subject = {Explainable Artificial Intelligence}, language = {en} } @misc{KochRiehle2024, author = {Koch, Christian and Riehle, Dirk}, title = {Schema Inference with Explainable AI for Data Engineering in Government Institutions}, doi = {10.34646/thn/ohmdok-1505}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-15059}, pages = {2 -- 6}, year = {2024}, abstract = {Data engineering is an integral part of the data science process. It comprises tasks such as data ingestion, data transformation, and data quality assurance. In order to fulfill these tasks, schema inference is an important capability. Its goal is to detect the structure of a dataset and to derive metadata on hierarchies, data types, etc. Artificial intelligence (AI) has the potential to automate schema inference and thus increase the efficiency of the data science process. However, as government institutions are subject to special regulations, explainability of AI models can be a mandatory requirement. Goal of this research protocol is to plan a systematic review of literature on schema inference with explainable AI (XAI) for data engineering in government institutions. This second version includes adjustments resulting from the first iteration of the review.}, subject = {Explainable Artificial Intelligence}, language = {en} } @misc{KochRiehle2023, author = {Koch, Christian and Riehle, Dirk}, title = {Schema Inference with Explainable AI for Data Engineering in Government Institutions}, doi = {10.34646/thn/ohmdok-1160}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-11609}, pages = {2 -- 6}, year = {2023}, abstract = {Data engineering is an integral part of the data science process. It comprises tasks such as data ingestion, data transformation, and data quality assurance. In order to fulfill these tasks, schema inference is an important capability. Its goal is to detect the structure of a dataset and to derive metadata on hierarchies, data types, etc. Artificial intelligence (AI) has the potential to automate schema inference and thus increase the efficiency of the data science process. However, as government institutions are subject to special regulations, explainability of AI models can be a mandatory requirement. Goal of this research protocol is to plan a systematic review of literature on schema inference with explainable AI (XAI) for data engineering in government institutions.}, subject = {Explainable Artificial Intelligence}, language = {en} } @misc{KochKogelheideLorenz2023, author = {Koch, Christian and Kogelheide, Lucie and Lorenz, Raphael}, title = {Post-Quantum Cryptography with Python and Linux}, series = {Towards Data Science}, journal = {Towards Data Science}, publisher = {Medium}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-10814}, pages = {7 (ungez{\"a}hlt)}, year = {2023}, abstract = {If we believe Edward Snowden, encryption is "the only true protection against surveillance". However, advances in quantum technology might endanger this safeguard. Our article discusses why quantum computing poses a threat to data security and what to do about it. Instead of a purely theoretical analysis, we build on code examples using Python, C, and Linux.}, subject = {Computersicherheit}, language = {en} } @misc{KochStadiBerle2023, author = {Koch, Christian and Stadi, Markus and Berle, Lukas}, title = {From Data Engineering to Prompt Engineering}, series = {Towards Data Science}, journal = {Towards Data Science}, publisher = {Medium}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-10829}, pages = {9 (ungez{\"a}hlt)}, year = {2023}, abstract = {Data engineering makes up a large part of the data science process. In CRISP-DM this process stage is called "data preparation". It comprises tasks such as data ingestion, data transformation and data quality assurance. In our article we solve typical data engineering tasks using ChatGPT and Python. By doing so, we explore the link between data engineering and the new discipline of prompt engineering.}, subject = {K{\"u}nstliche Intelligenz}, language = {en} } @misc{Hoevel2019, author = {H{\"o}vel, Wolfhard}, title = {Experiments with integer vectors}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-4815}, pages = {17}, year = {2019}, abstract = {The algorithm described in [ 1 ], [ 2 ] and [ 3 ] is originally intended for integer vectors. Therefore, the programming language Python is well suited to represent large integers. In the following two Python source codes „EightPoints03.txt" and „EightPoints03-graph.txt" an example is discribed. The algorithm is derived exclusively from position and displacement vectors. In this example eight points are pairwise coupled so that four bivectors emerge. Within the bivectors the points are displaced by the step width h. The bivectors are coupled by charges. For the displacement step width pull caused by the charges shall apply 1 << pull << h. This is described by the reciprocal coupling factor reci: h = reci * pull. If one chooses reci close to 1, the coupling is so strong that chaos can be caused. If reci >> 100 is chosen, the coupling between the bivectors is very weak, so that the interactions are very slow. [ 1 ] https://opus4.kobv.de/opus4-ohm/frontdoor/index/index/docId/34 [ 2 ] https://opus4.kobv.de/opus4-ohm/frontdoor/index/index/docId/126 [ 3 ] https://opus4.kobv.de/opus4-ohm/frontdoor/index/index/docId/253}, subject = {Python }, language = {en} } @misc{Hoevel2018, author = {H{\"o}vel, Wolfhard}, title = {Experiments with vectors}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-2537}, pages = {5 S.}, year = {2018}, abstract = {In the Euclidean space some bivectors are created. These can be linked together. As a result, more or less stable particles are generated by self-organization. These particles influence each other in many ways. They can merge, disintegrate or are positioned more or less close to each other. The executable program Particle.jar demonstrates the variety of dynamic behavior in which the particles interact.}, subject = {Vektor}, language = {en} } @misc{FriedrichGrombachKirschneretal.2017, author = {Friedrich, Stefanie and Grombach, Daniel and Kirschner, Sebastian and Kruczek, Alexander and Laur, Fenja and Semmelroth, Ronja and Schinnagel, Katharina and Vasel, Julian and Wegner, Moritz}, title = {Analytic success stories}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-2371}, pages = {20 S.}, year = {2017}, abstract = {The aim of the case study seminar was to identify analytic solutions and evaluate and compare them regarding multiple attributes. Different providers of analytic solutions from Gartner's BI quadrant were selected and more than 30 success stories are detailed in the appendix. The document has been solely created by the students of the course. Each analytic success story is characterized by the following attributes: - the actual analytical problem of the company - the goal for the analytical solution - the approach to realize the analytical solution - the technological basis for the solution - the project results with their respective benefits}, subject = {Massendaten}, language = {en} }