Listen
31 Ergebnisse
Bereich: Suchergebnisse
Publikation Optimized Computational Diabetes Prediction with Feature Selection Algorithms(2023) Li, Xi; Curiger, Michèle; Dornberger, Rolf; Hanne, Thomas04B - Beitrag KonferenzschriftPublikation Computational Intelligence in Logistik und Supply Chain Management(Springer Gabler, 2023) Hanne, Thomas; Dornberger, RolfPräsentiert den aktuellen Stand der Technik beim Einsatz von Computational Intelligence in der Lieferkette. Behandelt Probleme in den Bereichen Bestands- und Produktionsplanung, Scheduling, Transportplanung. Überprüft die verfügbare Software und Informationssysteme für jeden der behandelten Problembereiche.02 - MonographiePublikation Echtzeit Ressourcendisposition von Personal und Rollmaterial in der Eisenbahnbranche(Innosuisse, 2023) Ehrenthal, Joachim; Hanne, Thomas; Telesko, Rainer; Gachnang, PhillipZu wenig Personal und Rollmaterial, kurzfristig angesagte Arbeiten an der Infrastruktur mit den entsprechenden betrieblichen Behinderungen und Einschränkungen sowie kurzfristig auftretende Störungen prägen zurzeit die Berichterstattung über die Entwicklungen im öffentlichen Verkehr der Schweiz. Es ist absehbar, dass sich diese unbefriedigende Situation über eine längere Zeitspanne kaum massgeblich verbessern wird. Umso wichtiger ist es, vorhandene Ressourcen optimal einzusetzen und den zukünftigen Bedarf an Mitarbeitenden und Rollmaterial in den Griff zu kriegen. Die Fachhochschulen der Ostschweiz (OST) und der Nordwestschweiz FHNW entwickelten mit der Südostbahn (SOB), den luxemburgischen Eisenbahnen (CFL) und der Eisenbahn-Softwareherstellerin Qnamic eine zukunftsweisende Software zur Unterstützung der Eisenbahn-Disposition, um in Echtzeit über situationsspezifische Massnahmenpakete zur Störungsbehebung zu verfügen.05 - Forschungs- oder ArbeitsberichtPublikation FLIE with rules(2021) Pustulka, Elzbieta; Hanne, Thomas; de Espona, LucíaFLIE (Form Labelling for Information Extraction) allows us to extract information from Swiss insurance policies. Insurance policies are forms which are weakly aligned and do not lend themselves to automated data extraction without preprocessing. Our preprocessing annotates data with geometry and combined with manual training data generation gives the extraction accuracy of over 80% for a subset of attributes which have been seen 8 times or more. In this paper we extend FLIE with rules. The aim is to compare machine learning used in FLIE to the standard industry approach of using rules to extract data. We hand crafted rules (regular expressions in Python) for the KTG insurance (27 rules), UVG insurance (29 rules), and UVG-Z (23 rules), for each insurance type covering around 20 attributes. We also generated rules for building insurance policies which we were new to (16 rules encoded in SpaCy). In all cases we saw that using rules alone gives us a similar accuracy in data extraction to machine learning (around 80%). In the case of building insurance the accuracy is higher, above 96%, with precision and recall around 89-92%. To support annotation and experimental evaluation, we created an annotation GUI and a GUI which automates the ML experiment. Planned work includes a comparison of rule based and ML approaches and extension to further policy types.06 - PräsentationPublikation Improved long-short term memory U-Net for image segmentation(Springer, 2021) Oller, Heide; Dornberger, Rolf; Hanne, Thomas; Thampi, Sabu M.; Krishnan, Sri; Hegde, Rajesh M.; Ciuonzo, Domenico; Hanne, Thomas; Kannan R., Jagadeesh04B - Beitrag KonferenzschriftPublikation Parameter selection for ant colony optimization for solving the travelling salesman problem based on the problem size(2021) Kempter, Philipp; Schmitz, Martin Peter; Hanne, Thomas; Dornberger, Rolf; Abraham, Ajith; Hanne, Thomas; Castillo, Oscar; Gandhi, Niketa; Nogueira Rios, Tatiane; Hong, Tzung-Pei04B - Beitrag KonferenzschriftPublikation FLIE: form labeling for information extraction(2021) Pustulka, Elzbieta; Hanne, Thomas; Gachnang, Phillip; Biafora, Pasquale; Arai, Kohei; Kapoor, Supriya; Bhatia, RahulInformation extraction (IE) from forms remains an unsolved problem, with some exceptions, like bills. Forms are complex and the templates are often unstable, due to the injection of advertising, extra conditions, or document merging. Our scenario deals with insurance forms used by brokers in Switzerland. Here, each combination of insurer, insurance type and language results in a new document layout, leading to a few hundred document types. To help brokers extract data from policies, we developed a new labeling method, called FLIE (form labeling for information extraction). FLIE first assigns a document to a cluster, grouping by language, insurer, and insurance type. It then labels the layout. To produce training data, the user annotates a sample document by hand, adding attribute names, i.e. provides a mapping. FLIE applies machine learning to propagate the mapping and extracts information. Our results are based on 24 Swiss policies in German: UVG (mandatory accident insurance), KTG (sick pay insurance), and UVGZ (optional accident insurance). Our solution has an accuracy of around 84-89%. It is currently being extended to other policy types and languages.04B - Beitrag KonferenzschriftPublikation Artificial intelligence and machine learning for maturity evaluation and model validation(2022) Hanne, Thomas; Gachnang, Phillip; Gatziu Grivas, Stella; Kirecci, Ilyas; Schmitter, PaulIn this paper, we discuss the possibility of using machine learning (ML) to specify and validate maturity models, in particular maturity models related to the assessment of digital capabilities of an organization. Over the last decade, a rather large number of maturity models have been suggested for different aspects (such as type of technology or considered processes) and in relation to different industries. Usually, these models are based on a number of assumptions such as the data used for the assessment, the mathematical formulation of the model and various parameters such as weights or importance indicators. Empirical evidence for such assumptions is usually lacking. We investigate the potential of using data from assessments over time and for similar institutions for the ML of respective models. Related concepts are worked out in some details and for some types of maturity assessment models, a possible application of the concept is discussed.04B - Beitrag KonferenzschriftPublikation powerGhosts & defensiveGhosts – Enhanced ghost team controller based on Ant Colony Optimization for Ms. Pac-Man(IEEE, 2021) Applewhite, Timothy; Kaufmann, Roger; Dornberger, Rolf; Hanne, ThomasThis paper presents an improved controller based on Ant Colony Optimization for the ghost team of Ms. Pac-Man. The controller is an enhanced version of fairGhosts which is based on the ghostAnt framework. Various improvements were implemented in terms of parameters and concepts. Especially for the explorer ants, fairGhosts uses a simplified version, as the exact reasoning for the proposed concepts could not be determined in the ghostAnt framework. In this paper, two new types of ghost teams are proposed after the modifications were conducted: powerGhosts and defensiveGhosts. powerGhosts take into account the power pill aspect for stopping criteria and solution quality of the explorer ants, and defensiveGhosts additionally involve a threshold of Ms. Pac-Man’s distance to the nearest power pill for the hunter ants, so as not to be caught easily. Test results show that the powerGhosts version shows on average 15% better results than the initial fairGhosts setup, while defensiveGhosts performs equal or slightly worse than the initial implementation. It can be concluded that including the power pill aspect in the explorer ants concept shows an improved performance of the ghost team. On the other hand, the concept of distancing ghosts from Ms. Pac-Man when she is near a power pill did not result in any significant improvement.04B - Beitrag KonferenzschriftPublikation A serious game for teaching genetic algorithms(2021) Moser, Lars; Saner, Kevin; Oggier, Vincent; Hanne, Thomas; Arai, Kohei04B - Beitrag Konferenzschrift