Browsing by Subject "decision trees"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Global Translation of Machine Learning Models to Interpretable Models
    (2021-12) Almerri, Mohammad; Ben Miled, Zina; Christopher, Lauren; Salama, Paul
    The widespread and growing usage of machine learning models, especially in highly critical areas such as law, predicate the need for interpretable models. Models that cannot be audited are vulnerable to inheriting biases from the dataset. Even locally interpretable models are vulnerable to adversarial attack. To address this issue a new methodology is proposed to translate any existing machine learning model into a globally interpretable one. This methodology, MTRE-PAN, is designed as a hybrid SVM-decision tree model and leverages the interpretability of linear hyperplanes. MTRE-PAN uses this hybrid model to create polygons that act as intermediates for the decision boundary. MTRE-PAN is compared to a previously proposed model, TRE-PAN, on three non-synthetic datasets: Abalone, Census and Diabetes data. TRE-PAN translates a machine learning model to a 2-3 decision tree in order to provide global interpretability for the target model. The datasets are each used to train a Neural Network that represents the non-interpretable model. For all target models, the results show that MTRE-PAN generates interpretable decision trees that have a lower number of leaves and higher parity compared to TRE-PAN.
  • Thumbnail Image
    Item
    Performance of Response Based One Shot Controls Handling Missing Phasor Measurements
    (IEEE, 2020-08) Dahal, Niraj; Rovnyak, Steven M.; Electrical and Computer Engineering, School of Engineering and Technology
    With the advent of real-time PMU data acquisition technology, the possibility of solutions to several instability problems in power system has increased. However, PMUs may undergo different data quality issues like recording bad data or missing data. Some paper mentions about 5-10% of missing samples in some historical PMU's dataset. This paper assumes 0-10% of missing phasor samples by randomly deleting measurements and explores imputation methods of handling missing data in real time. The simulation is carried out in a DT-based stability prediction and one-shot control scheme of WECC's 176-bus model. Several control performances are evaluated to decide a useful method of missing data recovery for the response based one shot control scheme. A PMU data quality issue is not limited to missing samples only but also interference with noises. Later part of this paper performs simulation considering noisy phasor measurements. A 45 dB of Gaussian distributed noise is deliberately added to phasor samples and simulation is performed with different DT indices and thresholds for real time stability prediction and control actuation.