A Novel Ensemble Model - The Random Granular Reflections

2021 ◽  
Vol 179 (2) ◽  
pp. 183-203
Author(s):  
Piotr Artiemjew ◽  
Krzysztof Ropiak
Keyword(s):  
Random Forests ◽  
Learning Process ◽  
Ensemble Methods ◽  
Training System ◽  
New Method ◽  
Ensemble Model ◽  
Optimal Parameters ◽  
New Approach ◽  
Decision Systems

One of the most popular families of techniques to boost classification are Ensemble methods. Random Forests, Bagging and Boosting are the most popular and widely used ones. This article presents a novel Ensemble Model, named Random Granular Reflections. The algorithm used in this new approach creates an ensemble of homogeneous granular decision systems. The first step of the learning process is to take the training system and cover it with random homogeneous granules (groups of objects from the same decision class that are as little indiscernible from each other as possible). Next, granular reflection is created, which is finally used in the classification process. Results obtained by our initial experiments show that this approach is promising and comparable with other tested methods. The main advantage of our new method is that it is not necessary to search for optimal parameters while looking for granular reflections in the subsequent iterations of our ensemble model.

A New Approach for Assessing Learners in an Online Problem Based Learning Environment

2020 ◽  
pp. 307-324
Author(s):  
Houda Tadjer ◽  
Yacine Lafifi ◽  
Hassina Seridi-Bouchelaghem
Keyword(s):  
Problem Solving ◽  
Computer Science ◽  
Small Groups ◽  
Learning Process ◽  
Problem Based Learning ◽  
Point Of View ◽  
New Method ◽  
Science Department ◽  
New Approach ◽  
Problem Solving Process

Problem-based learning (PBL) is an approach that improves students' skills in problem solving. The application of PBL as an approach of teaching in a class requires students' presence; such constraint cannot be fulfilled by all students. Therefore, it is important to avoid this problem by implementing an online PBL environment where students are grouped remotely and work together to solve a problem proposed by the teacher. This will guide the learning process of the learners and can evaluate their solution. In reality, we can find members who do not really contribute to solve a problem. From this point of view, the teacher's evaluation will not be adequate to estimate the contribution of the learner in the solution of a given problem. Therefore, it is important to think of another way for assessing learners' solution. So, the challenge is to implement an online PBL environment and to propose a new method for assessing students. In this paper, the authors present their system called Problearn. The developed system allows students to solve problems remotely in small groups. Furthermore, the system evaluates each student based on his behavioral profiles during the problem-solving process. To do so, the system must keep track of different actions carried out by the students. This system has been tested by students of a computer science department where they achieved very good results.

scholarly journals Spectrum of Variable-Random Trees

2008 ◽  
Vol 32 ◽  
pp. 355-384 ◽  
Author(s):  
F. T. Liu ◽  
K. M. Ting ◽  
Y. Yu ◽  
Z. H. Zhou
Keyword(s):  
Random Forests ◽  
Ensemble Methods ◽  
Random Trees ◽  
Training Data ◽  
Random Subspace ◽  
Specific Level ◽  
New Approach ◽  
Diverse Ensemble ◽  
Base Learner ◽  
Better Than

In this paper, we show that a continuous spectrum of randomisation exists, in which most existing tree randomisations are only operating around the two ends of the spectrum. That leaves a huge part of the spectrum largely unexplored. We propose a base learner VR-Tree which generates trees with variable-randomness. VR-Trees are able to span from the conventional deterministic trees to the complete-random trees using a probabilistic parameter. Using VR-Trees as the base models, we explore the entire spectrum of randomised ensembles, together with Bagging and Random Subspace. We discover that the two halves of the spectrum have their distinct characteristics; and the understanding of which allows us to propose a new approach in building better decision tree ensembles. We name this approach Coalescence, which coalesces a number of points in the random-half of the spectrum. Coalescence acts as a committee of ``experts'' to cater for unforeseeable conditions presented in training data. Coalescence is found to perform better than any single operating point in the spectrum, without the need to tune to a specific level of randomness. In our empirical study, Coalescence ranks top among the benchmarking ensemble methods including Random Forests, Random Subspace and C5 Boosting; and only Coalescence is significantly better than Bagging and Max-Diverse Ensemble among all the methods in the comparison. Although Coalescence is not significantly better than Random Forests, we have identified conditions under which one will perform better than the other.

Janus – A New Approach to Air Combat Pilot Training

2019 ◽  
Vol 2019 (4) ◽  
pp. 7-22
Author(s):  
Georges Bridel ◽  
Zdobyslaw Goraj ◽  
Lukasz Kiszkowiak ◽  
Jean-Georges Brévot ◽  
Jean-Pierre Devaux ◽  
...  
Keyword(s):  
Low Cost ◽  
Cost Effective ◽  
High Energy ◽  
Training System ◽  
Embedded Sensors ◽  
Pilot Training ◽  
New Approach ◽  
Combat Aircraft ◽  
Air Combat ◽  
The Cost

Abstract Advanced jet training still relies on old concepts and solutions that are no longer efficient when considering the current and forthcoming changes in air combat. The cost of those old solutions to develop and maintain combat pilot skills are important, adding even more constraints to the training limitations. The requirement of having a trainer aircraft able to perform also light combat aircraft operational mission is adding unnecessary complexity and cost without any real operational advantages to air combat mission training. Thanks to emerging technologies, the JANUS project will study the feasibility of a brand-new concept of agile manoeuvrable training aircraft and an integrated training system, able to provide a live, virtual and constructive environment. The JANUS concept is based on a lightweight, low-cost, high energy aircraft associated to a ground based Integrated Training System providing simulated and emulated signals, simulated and real opponents, combined with real-time feedback on pilot’s physiological characteristics: traditionally embedded sensors are replaced with emulated signals, simulated opponents are proposed to the pilot, enabling out of sight engagement. JANUS is also providing new cost effective and more realistic solutions for “Red air aircraft” missions, organised in so-called “Aggressor Squadrons”.

A New Approach for the Analysis of Mixture Toxicity Data

1992 ◽  
Vol 26 (9-11) ◽  
pp. 2345-2348 ◽  
Author(s):  
C. N. Haas
Keyword(s):  
Quantitative Analysis ◽  
Mixture Toxicity ◽  
New Method ◽  
Metal Exposure ◽  
Positive Interactions ◽  
Toxicity Data ◽  
Data Set ◽  
New Approach ◽  
Negative Interactions

A new method for the quantitative analysis of multiple toxicity data is described and illustrated using a data set on metal exposure to copepods. Positive interactions are observed for Ni-Pb and Pb-Cr, with weak negative interactions observed for Ni-Cr.

scholarly journals New Method for Generating New Families of Distributions

Symmetry ◽  
2021 ◽  
Vol 13 (4) ◽  
pp. 726
Author(s):  
Lamya A. Baharith ◽  
Wedad H. Aljuhani
Keyword(s):  
Exponential Distribution ◽  
Real Data ◽  
Rate Function ◽  
New Method ◽  
Likelihood Method ◽  
Alpha Power ◽  
Unknown Parameters ◽  
Failure Rate Function ◽  
New Approach ◽  
Numerical Studies

This article presents a new method for generating distributions. This method combines two techniques—the transformed—transformer and alpha power transformation approaches—allowing for tremendous flexibility in the resulting distributions. The new approach is applied to introduce the alpha power Weibull—exponential distribution. The density of this distribution can take asymmetric and near-symmetric shapes. Various asymmetric shapes, such as decreasing, increasing, L-shaped, near-symmetrical, and right-skewed shapes, are observed for the related failure rate function, making it more tractable for many modeling applications. Some significant mathematical features of the suggested distribution are determined. Estimates of the unknown parameters of the proposed distribution are obtained using the maximum likelihood method. Furthermore, some numerical studies were carried out, in order to evaluate the estimation performance. Three practical datasets are considered to analyze the usefulness and flexibility of the introduced distribution. The proposed alpha power Weibull–exponential distribution can outperform other well-known distributions, showing its great adaptability in the context of real data analysis.

scholarly journals An Optimized Stacking Ensemble Model for Phishing Websites Detection

Electronics ◽  
2021 ◽  
Vol 10 (11) ◽  
pp. 1285
Author(s):  
Mohammed Al-Sarem ◽  
Faisal Saeed ◽  
Zeyad Ghaleb Al-Mekhlafi ◽  
Badiea Abdulkarem Mohammed ◽  
Tawfik Al-Hadhrami ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forests ◽  
Ensemble Method ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Ensemble Model ◽  
Security Attacks ◽  
Data Set ◽  
Machine Learning Methods ◽  
Ensemble Machine Learning

Security attacks on legitimate websites to steal users’ information, known as phishing attacks, have been increasing. This kind of attack does not just affect individuals’ or organisations’ websites. Although several detection methods for phishing websites have been proposed using machine learning, deep learning, and other approaches, their detection accuracy still needs to be enhanced. This paper proposes an optimized stacking ensemble method for phishing website detection. The optimisation was carried out using a genetic algorithm (GA) to tune the parameters of several ensemble machine learning methods, including random forests, AdaBoost, XGBoost, Bagging, GradientBoost, and LightGBM. The optimized classifiers were then ranked, and the best three models were chosen as base classifiers of a stacking ensemble method. The experiments were conducted on three phishing website datasets that consisted of both phishing websites and legitimate websites—the Phishing Websites Data Set from UCI (Dataset 1); Phishing Dataset for Machine Learning from Mendeley (Dataset 2, and Datasets for Phishing Websites Detection from Mendeley (Dataset 3). The experimental results showed an improvement using the optimized stacking ensemble method, where the detection accuracy reached 97.16%, 98.58%, and 97.39% for Dataset 1, Dataset 2, and Dataset 3, respectively.

Application of an Improved Colored Petri Net Modeling Method in Certain Missile Training System

2015 ◽  
Vol 742 ◽  
pp. 330-334
Author(s):  
Chun Jian Wang ◽  
Wei Yue ◽  
Hai Yan Ji
Keyword(s):  
Complex System ◽  
Petri Net ◽  
Time Use ◽  
Training System ◽  
New Method ◽  
Mathematics Model ◽  
Research Approach ◽  
Complex Environments ◽  
Training Simulator ◽  
First Time

In allusion to the need of analyzing complex system, we have proposed a method named multi-grade color Petri net. We for the first time use this new method to analyze a missile training simulator system. This model can accurately reflect the complex environments of the system and avoid the difficulty occurring often in developing accurate mathematics model by using classical research approach.

Comparison of New Telescopic Video Microlaryngoscopic and Standard Microlaryngoscopic Techniques

1992 ◽  
Vol 101 (1) ◽  
pp. 51-60 ◽  
Author(s):  
Eiji Yanagisawa ◽  
Ken Yanagisawa ◽  
Jay B. Horowitz ◽  
Lawrence J. Mambrino
Keyword(s):  
Video Camera ◽  
New Method ◽  
Depth Of Field ◽  
Microscopic Technique ◽  
New Approach ◽  
Advantages And Disadvantages ◽  
Laryngeal Surgery ◽  
Valuable Method ◽  
Microlaryngeal Surgery ◽  
Television Image

A new approach to microlaryngeal surgery using a specially designed video microlaryngoscope with a rigid endoscopic telescope and an attached video camera was introduced by Kantor et al in 1990. The ability to video document and perform surgery of the larynx by viewing a high-resolution television image was demonstrated. This method was recommended over the standard microscopic technique for increased visibility with greater depth of field, unimpeded instrument access, instant documentation, and superior teaching value. The authors tried this new method and the standard microscopic technique at the same sitting on a series of patients. This paper will compare these two different techniques and discuss their advantages and disadvantages. Although the new method has many advantages, the standard microscopic technique remains as a valuable method in laryngeal surgery.

Inferring age from otolith measurements: a review and a new approach

2004 ◽  
Vol 61 (7) ◽  
pp. 1269-1284 ◽  
Author(s):  
RIC Chris Francis ◽  
Steven E Campana
Keyword(s):  
Simulation Experiment ◽  
Cost Benefit Analysis ◽  
Cost Benefit ◽  
New Method ◽  
Population Parameters ◽  
Benefit Analysis ◽  
Mixture Analysis ◽  
New Approach ◽  
Individual Fish ◽  
The Cost

In 1985, Boehlert (Fish. Bull. 83: 103–117) suggested that fish age could be estimated from otolith measurements. Since that time, a number of inferential techniques have been proposed and tested in a range of species. A review of these techniques shows that all are subject to at least one of four types of bias. In addition, they all focus on assigning ages to individual fish, whereas the estimation of population parameters (particularly proportions at age) is usually the goal. We propose a new flexible method of inference based on mixture analysis, which avoids these biases and makes better use of the data. We argue that the most appropriate technique for evaluating the performance of these methods is a cost–benefit analysis that compares the cost of the estimated ages with that of the traditional annulus count method. A simulation experiment is used to illustrate both the new method and the cost–benefit analysis.

Sign in / Sign up

Export Citation Format

Share Document