Informatik

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 5 of 172
  • Item
    Using Learning Analytics to Identify Student Learning Profiles for Software Development Courses
    (New York, NY : Association for Computing Machinery, 2023) Söchtig, Philipp; Apel, Sebastian; Windisch, Hans-Michael; Mottok, Jürgen
    Often lecturers encounter the problem of not knowing how students use the course materials during a semester. In our approach we devised a web-based system that presents all learning materials in a digital format, allowing us to record student learning activities. The recorded usage data enabled extensive analyses of student learning behaviour which can support lecturers with improving the materials as well as understanding students’ learning material preferences and learning profiles, which can be composed by combining different usage modes depending on the material used. For the lectures we analysed, a higher success in the exam can be correlated to higher usage of the learning material according to our research data. Furthermore, student preferences regarding the form of presentation (f.e. slides over videos) could also be seen.
  • Item
    In-situ fruit analysis by means of LiDAR 3D point cloud of normalized difference vegetation index (NDVI)
    (Amsterdam [u.a.] : Elsevier, 2023) Tsoulias, Nikos; Saha, Kowshik Kumar; Zude-Sasse, Manuela
    A feasible method to analyse fruit at the tree is requested in precise production management. The employment of light detection and ranging (LiDAR) was approached aimed at measuring the number of fruit, quality-related size, and ripeness-related chlorophyll of fruit skin. During fruit development (65 – 130 day after full bloom, DAFB), apples were harvested and analysed in the laboratory (n = 225) with two LiDAR laser scanners measuring at 660 and 905 nm. From these two 3D point clouds, the normalized difference vegetation index (NDVILiDAR) was calculated. The correlation analysis of NDVILiDAR and chemically analysed fruit chlorophyll content showed R2 = 0.81 and RMSE = 3.63 % on the last measuring date, when fruit size reached 76 mm. The method was tested on 3D point clouds of 12 fruit trees measured directly in the orchard, during fruit growth on five measuring dates, and validated with manual fruit analysis in the orchard (n = 4632). Point clouds of individual apples were segmented from 3D point clouds of trees and fruit NDVILiDAR were calculated. The non-invasively obtained field data showed good calibration performance capturing number of fruit, fruit size, fruit NDVILiDAR, and chemically analysed chlorophyll content of R2 = 0.99, R2 = 0.98 with RMSE = 3.02 %, R2 = 0.65 with RMSE = 0.65 %, R2 = 0.78 with RMSE = 1.31 %, respectively, considering the related reference data at last measuring date 130 DAFB. The new approach of non-invasive laser scanning provided physiologically and agronomically valuable time series data on differences in fruit chlorophyll affected by the leaf area to number of fruit and leaf area to fruit fresh mass ratios. Concluding, the method provides a tool for gaining production-relevant plant data for, e.g., crop load management and selective harvesting by harvest robots.
  • Item
    Inverse learning in Hilbert scales
    (Dordrecht [u.a.] : Springer Science + Business Media B.V, 2023) Rastogi, Abhishake; Mathé, Peter
    We study linear ill-posed inverse problems with noisy data in the framework of statistical learning. The corresponding linear operator equation is assumed to fit a given Hilbert scale, generated by some unbounded self-adjoint operator. Approximate reconstructions from random noisy data are obtained with general regularization schemes in such a way that these belong to the domain of the generator. The analysis has thus to distinguish two cases, the regular one, when the true solution also belongs to the domain of the generator, and the ‘oversmoothing’ one, when this is not the case. Rates of convergence for the regularized solutions will be expressed in terms of certain distance functions. For solutions with smoothness given in terms of source conditions with respect to the scale generating operator, then the error bounds can then be made explicit in terms of the sample size.
  • Item
    Redox Memristors with Volatile Threshold Switching Behavior for Neuromorphic Computing
    (Windsor ; Beijing : English China Online Journals, ECOJ, 2022) Wang, Yu-Hao; Gong, Tian-Cheng; Ding, Ya-Xin; Li, Yang; Wang, Wei; Chen, Zi-Ang; Du, Nan; Covi, Erika; Farronato, Matteo; Ielmini, Daniele; Zhang, Xu-Meng; Luo, Qing
    The spiking neural network (SNN), closely inspired by the human brain, is one of the most powerful platforms to enable highly efficient, low cost, and robust neuromorphic computations in hardware using traditional or emerging electron devices within an integrated system. In the hardware implementation, the building of artificial spiking neurons is fundamental for constructing the whole system. However, with the slowing down of Moore’s Law, the traditional complementary metal-oxide-semiconductor (CMOS) technology is gradually fading and is unable to meet the growing needs of neuromorphic computing. Besides, the existing artificial neuron circuits are complex owing to the limited bio-plausibility of CMOS devices. Memristors with volatile threshold switching (TS) behaviors and rich dynamics are promising candidates to emulate the biological spiking neurons beyond the CMOS technology and build high-efficient neuromorphic systems. Herein, the state-of-the-art about the fundamental knowledge of SNNs is reviewed. Moreover, we review the implementation of TS memristor-based neurons and their systems, and point out the challenges that should be further considered from devices to circuits in the system demonstrations. We hope that this review could provide clues and be helpful for the future development of neuromorphic computing with memristors.
  • Item
    Text-to-Ontology Mapping via Natural Language Processing with Application to Search for Relevant Ontologies in Catalysis †
    (Basel : MDPI, 2023) Korel, Lukáš; Yorsh, Uladzislau; Behr, Alexander S.; Kockmann, Norbert; Holeňa, Martin
    The paper presents a machine-learning based approach to text-to-ontology mapping. We explore a possibility of matching texts to the relevant ontologies using a combination of artificial neural networks and classifiers. Ontologies are formal specifications of the shared conceptualizations of application domains. While describing the same domain, different ontologies might be created by different domain experts. To enhance the reasoning and data handling of concepts in scientific papers, finding the best fitting ontology regarding description of the concepts contained in a text corpus. The approach presented in this work attempts to solve this by selection of a representative text paragraph from a set of scientific papers, which are used as data set. Then, using a pre-trained and fine-tuned Transformer, the paragraph is embedded into a vector space. Finally, the embedded vector becomes classified with respect to its relevance regarding a selected target ontology. To construct representative embeddings, we experiment with different training pipelines for natural language processing models. Those embeddings in turn are later used in the task of matching text to ontology. Finally, the result is assessed by compressing and visualizing the latent space and exploring the mappings between text fragments from a database and the set of chosen ontologies. To confirm the differences in behavior of the proposed ontology mapper models, we test five statistical hypotheses about their relative performance on ontology classification. To categorize the output from the Transformer, different classifiers are considered. These classifiers are, in detail, the Support Vector Machine (SVM), k-Nearest Neighbor, Gaussian Process, Random Forest, and Multilayer Perceptron. Application of these classifiers in a domain of scientific texts concerning catalysis research and respective ontologies, the suitability of the classifiers is evaluated, where the best result was achieved by the SVM classifier.