KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Subject: search.filters.subject.Ridge regression

Search Results

Now showing 1 - 3 of 3
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Machine Learning-Based Prediction of Undrained Shear Strength in Marine Alluvial Clays: A Case Study of Bangkok
    (2026-01-01)
    Ramineni, Sai Krishna Akash
    ;
    Song, Zejun
    ;
    Garg, Ankit
    ;
    Kamchoom, Viroon
    Accurate evaluation of undrained shear strength (Su) is crucial for the safe design of foundations and slopes in marine alluvial clays, including those commonly found in Bangkok. In this study, we assembled an automated machine learning (AutoML) workflow using open-source Python libraries to explore suitable predictive models for Su based on 152 undisturbed clay samples. The input variables considered include depth, moisture content, liquid limit, plastic limit, vane shear strength (PP), and total unit weight. Across the models evaluated, ridge regression offered a stable balance between accuracy and computational efficiency, with a mean absolute error of 0.550 t/m<sup>2</sup>, a root mean square error of 0.710 t/m<sup>2,</sup>, and an R² of 0.809, while requiring less than 0.05s of training time. The AutoML process facilitated a more transparent comparison of candidate algorithms, providing insight into variable relevance. Specifically, PP, depth, and unit weight emerged as the most influential predictors. Traditional index properties showed comparatively lower contributions. Five-fold cross-validation suggested that the selected model maintained consistent performance (mean R² = 0.810; standard deviation = 0.025). These results suggest that a streamlined AutoML workflow can aid in identifying reliable and easy-to-interpret models for Su estimation in Bangkok clays. Such an approach may complement laboratory testing and help reduce some of the uncertainty associated with empirical correlations, especially in preliminary design stages.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Comparing penalized regression analysis of logistic regression model with multicollinearity
    (2019-07-08)
    Araveeporn, Autcha
    ;
    Kuharatanachai, Choojai
    The goal of this research is to estimate the parameter of the logistic regression model by penalized regression analysis which consisted of ridge regression, lasso, and elastic net method. The logistic regression is considered between a binary dependent variable and 3 and 5 independent variables. The independent variables are generated from normal distribution, contaminated normal distribution, and t distribution on correlation coefficient at 0.1, 0.5, and 0.99 or called multicollinearity problem. The maximum likelihood estimator has used as the classical method by differential the log likelihood function with respect to the coefficients. Ridge regression is to choose the unknown ridge parameter by cross-validation, so ridge estimator is evaluated by the adding ridge parameter on penalty term. Lasso (least absolute shrinkage and selection operator) is added the penalty term on scales sum of the absolute value of the coefficients. The elastic net can be mixed between ridge regression and lasso on the penalty term. The criterion of these methods is compared by percentage of predicted accuracy value. The results are found that lasso is satisfied when the independent variables are simulated from normal and t distribution in most cases, and the lasso outperforms on the contaminated normal distribution.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Hinge loss projection for classification
    (2016-01-01)
    Alfarozi, Syukron Abu Ishaq
    ;
    Woraratpanya, Kuntpong
    ;
    Pasupa, Kitsuchart
    ;
    Sugimoto, Masanori
    Hinge loss is one-sided function which gives optimal solution than that of squared error (SE) loss function in case of classification. It allows data points which have a value greater than 1 and less than −1 for positive and negative classes, respectively. These have zero contribution to hinge function. However, in the most classification tasks, least square (LS) method such as ridge regression uses SE instead of hinge function. In this paper, a simple projection method is used to minimize hinge loss function through LS methods. We modify the ridge regression and its kernel based version i.e. kernel ridge regression so that it can adopt to hinge function instead of using SE in case of classification problem. The results show the effectiveness of hinge loss projection method especially on imbalanced data sets in terms of geometric mean (GM).