Logo des Repositoriums

Auflistung nach Schlagwort "feature extraction"

1 - 2 von 2
  • Konferenzbeitrag
    Independent components analysis of starch deficient pgm mutants
    (German Conference on Bioinformatics 2004, GCB 2004, 2004) Scholz, Matthias; Gibon, Yves; Stitt, Mark; Selbig, Joachim
    Changes in enzymatic activities in response to carbon starvation were investigated in Arabidopsis thaliana in two distinct experiments. One compares the Columbia ecotype (Col-0) and its starch deficient pgm mutant (plastidial phosphoglucomutase), the other investigates the enzymatic activities of Col-0 under extended night conditions. A classical technique for detecting and visualizing relevant information from the measured data is principal component analysis (PCA). We show that independent component analysis (ICA) is more suitable for our questions and the results are more precise than those obtained with PCA. This higher informative power is only achieved when ICA is combined with suitable pre-processing and evaluation criteria. It is essential to first reduce the dimensionality of the data set, using PCA. The number of principal components determines the quality of ICA significantly, therefore we propose a criterion for estimating the optimal dimension automatically. The measure of kurtosis is used to sort the extracted components. We found that ICA could detect on the one hand the time component of the extended night experiment, and on the other hand a discriminating component in the pgm mutant experiment. In both components the most important enzymes were the same, confirming the carbon starvation phenotype in the mutant.
  • Konferenzbeitrag
    Screen me, Smartphone! Using an AI-Screening Tool to Assists Underage Refugees in Recognizing Potential Traumatization
    (Mensch und Computer 2021 - Tagungsband, 2021) Mühl, Lisa; Horstmann, Aike C.; Wittenborn, André; Storch, Dunja; Krajewski, Jarek
    Many refugees experience critical life events or traumatic injuries during their flights. Here, underaged, (un)accompanied refugees are a particularly vulnerable group. To date, there are insufficient support structures that recognize the specific demands and allow for careful and early identification of indicators of traumatization or behavioral problems. As an approach to counteract these deficits and support underage refugees, the TraM project investigates the potential of an AI-based screening tool providing indications of post-traumatic stress disorder via speech-emotion-recognition. A data collection for standardized learning data was conducted as a basis for the described screening module and the planned algorithms for automatic classification. We encountered several challenges such as insufficient data quality, uncertain classifications, and comorbidities such as depression as potentially confounding factors. Accordingly, emphasis lays on using the screening module for an initial examination of mental health and potential traumatization. This may encourage affected underage refugees to seek the help that is often highly needed.