Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 32 for “"Tree-based algorithms"”.
-
Novel tree-based algorithms for computational electromagnetics
Tree-based methods have wide applications for solving large-scale problems in electromagnetics, astrophysics, quantum chemistry, fluid mechanics, acoustics, and many more areas. This thesis focuses on their applicability for solving large-scale problems in electromagnetics. The Barnes-Hut (BH) …
-
Heat flow modeling with tree-based algorithms
… The third case study introduces a clustering-based Mixture-of-Experts framework. We developed a framework that partitions the input features into geophysically coherent subsets. Within each cluster, dedicated QRF models quantified both aleatoric and epistemic uncertainty, thereby improving …
-
New tree-based algorithms for network spare capacity design
… survivability techniques and the corresponding algorithms. Subsequently; we introduce our two pre-planned path restoration algorithms for spare capacity design in mesh-like networks. They can get higher utilization of spare capacity and reasonable reaction time. First one is a spanning tree …
-
Extending paratreet, a framework for spatial tree based algorithms, with gpu kernels and load balancers
Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-11 without embargo terms
-
Improving fairness of decision tree-based classification algorithms
The increasing use of machine learning algorithms in decision-making systems has led to certain concerns about fairness. Although machine learning algorithms, particularly tree-based classification algorithms, have generally been successful in efficiently handling large datasets, they are prone to …
-
Feature selection and hierarchical modelling in tree-based machine learning models.
Tree-based algorithms are quite popular in the machine learning area in general, due to its many advantages: interpretability, flexibility, high prediction power, and so on. They can be used to many different classification and regression problems, and are in constant development. Because of that, …
-
Frequent pattern mining of uncertain data streams
… patterns from streams of uncertain data, algorithms have been developed to use the sliding window model for processing and mining data streams. However, for some applications, the landmark window model and the time-fading model are more appropriate. In this M.Sc. thesis, I propose …
-
FREQUENCY-BASED FEATURE EXTRACTION FOR MALWARE CLASSIFICATION
Traditional signature-based malware detection is effective, but it can only identify known malicious programs. This thesis attempts to use machine-learning techniques to successfully identify previously unknown malware from a set of Windows executable programs. We analyzed the histogram of 4-, 8-, …
-
New Longitudinal Classification Approaches and Applications to Age-Related Disease Data
… the years. Regarding supervised learning algorithms, we focus on decision tree-based algorithms, namely Random Forests (which learn an ensemble of decision trees) and a decision tree algorithm (which learns a single decision tree). These algorithms were chosen because they represent a good …
-
Initial detailed routing algorithms
… synthesis flow. We study the fundamental routing algorithms such as maze routing, A*, and Steiner tree-based algorithms, as well as some global routing algorithms, namely FastRoute 4.0 and BoxRouter 2.0. We dissect some of the major state of the art initial detailed routing tools, such as …
-
Credit union profitability estimation and prediction: the case of lithuania /
… (ARDL). Estimated panel ARDL model, regression trees and more novel tree-based algorithms have been applied to predict future profitability ratios. The results indicate that country economic activity, average loan interest rates of banks and ratio of operating expenses to total income are …
-
Efficient frequent pattern mining from big data and its applications
… of many researchers. Consequently, many algorithms have been proposed. Popular algorithms include level-wise Apriori based algorithms, tree based algorithms, and hyperlinked array structure based algorithms. While these algorithms are popular and beneficial due to some nice properties, …
-
Predicting Spin-Symmetry Breaking in Organic Photovoltaic Compounds Using a Data Mining Approach
… learning classifier models (utilizing decision tree, random forest, neural network approaches) for this purpose, trained and tested on the molecular systems contained in the Harvard Clean Energy Project database. Our study employs Morgan fingerprints and molecular descriptors of the Dragon …
-
Large-scale non-linear prediction with applications
… large data sets. Several standard non-linear algorithms based on either kernels (e.g., kernel support vector machines and kernel ridge regression) or decision trees (e.g., gradient boosted decision trees) often yield superior predictive performance on various machine learning tasks compared to …
-
An integrated machine learning approach for predicting the quantity and quality of mine waste rock drainage
… In this research, multiple machine learning algorithms were integrated to predict the quantity and quality of waste rock drainage with regional weather data as the inputs: a tree-based algorithm to predict the occurrence of spring freshet, and two recurrent neural network algorithms to …
-
Convergence of a Reinforcement Learning Algorithm in Continuous Domains
… actions have been well studied, and there exist algorithms capable of producing a sequence of policies which converge to an optimal policy with probability one. Convergence guarantees for problems with continuous states also exist. Until recently, no online algorithm for continuous states and …
-
Reconfiguration of power networks based on graph-theoretic algorithms
… and implementing efficient graph-theoretic algorithms to analyze and solve such two real-world problems in power systems as follows.</p> <p>In response to disturbances, a self-healing system reconfiguration that splits a power network into self-sufficient islands can stop the propagation of …
-
Machine Learning-Driven Uncertainty Quantification and Parameter Analysis in Fire Risk Assessment for Nuclear Power Plants
… (ML) models, particularly neural networks and tree-based algorithms, with uncertainty quantification (UQ) techniques to enhance fire modeling and risk assessment in NPPs. The framework is applied to electrical enclosure cabinets and spill fires that represent about 50% of challenging fire …
-
Use of Machine Learning Methods to Predict the Unsteady Wind Pressure on Tall Buildings: An Integrated Approach
… this gap by developing a machine learning-based surrogate model to reconstruct unsteady wind pressure on generalised rectangular tall buildings considering spatial coordinates, wind direction (0°-90°), slenderness ratios (6-10) and side ratios (1-2). Boundary layer wind tunnel experiments …
-
Enhancing Data Science Job Market Transparency: Salary Prediction and Skill Valuation
… multiple model families—including linear models, tree-based algorithms, and multilayer perceptrons— to create an optimized ensemble model. Our best model achieves a Mean Absolute Percentage Error (MAPE) of 16%, demonstrating strong predictive performance.</p> <p>The second essay introduces a …
Page 1 of 2