Search results for " Methods"
showing 10 items of 4102 documents
International Education Studies: Increasing Their Linguistic Comparability by Developing Judgmental Reviews
2012
In international education studies, the different-language test versions need to be equally difficult to read and answer for the test to be valid. To ensure comparability, several quality control procedures have been developed. Among these, surprisingly little attention has been paid to judgmental reviews and their ability to identify language-related sources of bias. Also, the reviews have often failed in identifying biases. This paper explored whether it is possible to improve the ability of judgmental reviews to identify language-related sources of bias. A new review was made of two Finnish items which in the PISA (Programme for International Student Assessment) 2000 reading test showed …
RTS2 - the Remote Telescope System
2010
RTS2 is an open source observatory manager. It was written from scratch in the C++ language, with portability and modularity in mind. Its driving requirements originated from quick follow-ups of Gamma Ray Bursts. After some years of development it is now used to carry tasks it was originally not intended to carry. This article presents the current development status of the RTS2 code. It focuses on describing strategies which worked as well as things which failed to deliver expected results. Copyright © 2010 Petr Kubánek.
Artificial intelligence techniques for cancer treatment planning
1988
An artificial intelligence system, NEWCHEM, for the development of new oncology therapies is described. This system takes into account the most recent advances in molecular and cellular biology and in cell-drug interaction, and aims to guide experimentation in the design of new optimal protocols. Further work is being carried out, aimed to embody in the system all the basic knowledge of biology, physiopathology and pharmacology, to reason qualitatively from first principles so as to be able to suggest cancer therapies.
An ontology for cognitive mimetics
2018
AI and autonomous systems are intended to replace people in several jobs. People have worked in these jobs being able to execute the required information processing. This implies that new technical artefacts must be able to perform equitably effective information processing. Thus, it makes sense to develop the analysis of human information processing in designing intelligent systems. This approach has been termed cognitive mimetics. This paper studies how it would be practical to gain knowledge about human information processing and organize this knowledge using ontologies.
Questions in Cognitive Mimetics
2021
Human thinking advances through questions and answers. Any field of human endeavor is permeated by the presence of questions, answers and presuppositions. Questions have a kind of universality, whereby one can place the question marks on anything, including questions themselves. The process of asking the right questions about the right things and in the right way are key for the explication of an approach. Recently, we have begun thinking about an approach to the design of intelligent technology: Cognitive mimetics. In brief, the idea is to take inspiration of empirical human thinking in specific contexts to develop AI solutions. The purpose of this article is to question this approach from…
Multi-layer intrusion detection system with ExtraTrees feature selection, extreme learning machine ensemble, and softmax aggregation
2019
Abstract Recent advances in intrusion detection systems based on machine learning have indeed outperformed other techniques, but struggle with detecting multiple classes of attacks with high accuracy. We propose a method that works in three stages. First, the ExtraTrees classifier is used to select relevant features for each type of attack individually for each (ELM). Then, an ensemble of ELMs is used to detect each type of attack separately. Finally, the results of all ELMs are combined using a softmax layer to refine the results and increase the accuracy further. The intuition behind our system is that multi-class classification is quite difficult compared to binary classification. So, we…
Development of artificial neural network for condition assessment of bridges based on hybrid decision making method – Feasibility study
2021
Abstract Managing a bridge at an appropriate level of reliability requires knowledge of its technical condition, which is decisive in terms of maintenance and repair activities. This is a multi-criteria decision-making problem which results from the need to allocate limited financial resources to this work. Although many calculation models have been suggested in published sources, none of them has ever met these requirements. The algorithm presented by the authors allows for the assessment of any number of bridges, taking into account the diversity of solutions in terms of materials and structures, and can provide a solution to this problem. This hybrid calculation model, combining the modi…
Classical Training Methods
2006
This chapter reviews classical training methods for multilayer neural networks. These methods are widely used for classification and function modelling tasks. Nevertheless, they show a number of flaws or drawbacks that should be addressed in the development of such systems. They work by searching the minimum of an error function which defines the optimal behaviour of the neural network. Different standard problems are used to show the capabilities of these models; in particular, we have benchmarked the algorithms in a nonlinear classification problem and in three function modelling problems.
Neural network prediction in a system for optimizing simulations
2002
Neural networks have been widely used for both prediction and classification. Back-propagation is commonly used for training neural networks, although the limitations associated with this technique are well documented. Global search techniques such as simulated annealing, genetic algorithms and tabu search have also been used for this purpose. The developers of these training methods, however, have focused on accuracy rather than training speed in order to assess the merit of new proposals. While speed is not important in settings where training can be done off-line, the situation changes when the neural network must be trained and used on-line. This is the situation when a neural network i…
Estimation of Granger causality through Artificial Neural Networks: applications to physiological systems and chaotic electronic oscillators
2021
One of the most challenging problems in the study of complex dynamical systems is to find the statistical interdependencies among the system components. Granger causality (GC) represents one of the most employed approaches, based on modeling the system dynamics with a linear vector autoregressive (VAR) model and on evaluating the information flow between two processes in terms of prediction error variances. In its most advanced setting, GC analysis is performed through a state-space (SS) representation of the VAR model that allows to compute both conditional and unconditional forms of GC by solving only one regression problem. While this problem is typically solved through Ordinary Least Sq…