Krzysztof J. Cios, Ning Liu and Lucy S. Goodenday
A learning algorithm called CLILP2 (Cover Learning Using Integer Linear Programming) is applied to medical data to generate rules to recognize patients with coronary artery…
Abstract
A learning algorithm called CLILP2 (Cover Learning Using Integer Linear Programming) is applied to medical data to generate rules to recognize patients with coronary artery disease. The algorithm partitions a data set into subsets using features which best describe and distinguish a particular subset from all other subsets. These features are used to form the rules which can be used as the knowledge base of a diagnostic expert system. Results from the application of the algorithm to coronary artery stenosis data are compared with the results obtained from the existing expert system.
Details
Keywords
Krzysztof J. Cios and Ning Liu
Presents an inductive machine learning algorithm, CLILP2 that learns multiple covers for a concept from positive and negative examples. Although inductive learning is an…
Abstract
Presents an inductive machine learning algorithm, CLILP2 that learns multiple covers for a concept from positive and negative examples. Although inductive learning is an error‐prone process, multiple meaning interpretation of the examples is utilized by CLILP2 to compensate for the narrowness of induction. The algorithm is tested on data sets representing three different domains. The complexity of the algorithm is analysed and the results are compared with those obtained by others. Employs measures of specificity, sensitivity, and predictive accuracy not usually used in presenting machine learning results, and shows that they evaluate better the “correctness” of the learned concepts. Published in two parts: I – The CLILP2 algorithm; II – Experimental results and conclusions.
Details
Keywords
Krzysztof J. Cios and Ning Liu
Presents an inductive machine learning algorithm called CLILP2 that learns multiple covers for a concept from positive and negative examples. Although inductive learning is an…
Abstract
Presents an inductive machine learning algorithm called CLILP2 that learns multiple covers for a concept from positive and negative examples. Although inductive learning is an error‐prone process, multiple meaning interpretation of the examples is utilized by CLILP2 to compensate for the narrowness of induction. The algorithm is tested on data sets representing three different domains. Analyses the complexity of the algorithm and compares the results with those obtained by others. Employs measures of specificity, sensitivity, and predictive accuracy which are not usually used in presenting machine learning results, and shows that they evaluate better the “correctness” of the learned concepts. The study is published in two parts: I – the CLILP2 algorithm; II – experimental results and conclusions.
Details
Keywords
Krzysztof J. Cios and Ian Moraes
ALFS is an inductive learning algorithm that employs feature selection to learn concepts from examples. Features which best represent and differentiate a subset from other subsets…
Abstract
ALFS is an inductive learning algorithm that employs feature selection to learn concepts from examples. Features which best represent and differentiate a subset from other subsets in learning data are detected and used to produce rules. These rules form a knowledge base for an expert system. The performance of ALFS is illustrated using data sets from the domains of primary tumour and game playing.
Details
Keywords
Krzysztof J. Cios, Daniel K. Wedding and Ning Liu
Presents an inductive machine learning algorithm called CLIP3 (Cover learning using integer programming). CLIP3 is an extension of the CLILP2 algorithm. CLIP3 generates multiple…
Abstract
Presents an inductive machine learning algorithm called CLIP3 (Cover learning using integer programming). CLIP3 is an extension of the CLILP2 algorithm. CLIP3 generates multiple rules for a given concept from two sets of discrete attribute data. It combines the best concepts of tree‐based and rule‐based algorithms to produce a highly reliable machine‐learning algorithm. The algorithm is run on the benchmark “MONK′s data sets”. Compares the results of standard machine learning algorithms such as the ID and AQ families of algorithms. The algorithm is also run on the breast cancer data set and the results are compared with C4.5 algorithm results.
Details
Keywords
Patricia Ahmed and Rebecca Jean Emigh
Two perspectives provide alternative insights into household composition in contemporary Eastern Europe. The first stresses that individuals have relatively fixed preferences…
Abstract
Two perspectives provide alternative insights into household composition in contemporary Eastern Europe. The first stresses that individuals have relatively fixed preferences about living arrangements and diverge from them only when they cannot attain their ideal. The second major approach, the adaptive strategies perspective, predicts that individuals have few preferences. Instead, they use household composition to cope with economic hardship, deploy labor, or care for children or the elderly. This article evaluates these approaches in five post‐socialist East‐European countries, Bulgaria, Hungary, Poland, Romania, and Russia, using descriptive statistics and logistic regression. The results suggest that household extension is common in these countries and provide the most evidence for the adaptive strategies perspective. In particular, the results show that variables operationalizing the adaptive strategies perspective, including measures of single motherhood, retirement status, agricultural cultivation, and poverty, increase the odds of household extension.
Details
Keywords
Kamil Janeczek, Aneta Arazna and Krzysztof Lipiec
The aim of this paper is to present thermal and mechanical durability of conductive tracks screen-printed with silver polymer pastes on flexible magnetic sheets.
Abstract
Purpose
The aim of this paper is to present thermal and mechanical durability of conductive tracks screen-printed with silver polymer pastes on flexible magnetic sheets.
Design/methodology/approach
A test pattern that consisted of three straight lines was printed with two different silver pastes on a flexible magnetic sheet and a polyethylene naphthalate (PEN) foil for comparison. Electrical properties of these lines were examined by resistance measurements and their thickness was measured with a digital microscope on cross sections. Cyclic bending was performed to investigate mechanical properties of prepared samples as well as thermal shocks to analyse their thermal durability. Further, samples after thermal shocks underwent cyclic bending to test influence of thermal exposure on mechanical properties of the prepared samples. Changes in the test lines after the thermal and mechanical tests were assessed by resistance measurements and microscopic analysis of surface and internal structure of the test lines.
Findings
It was found that the most important factor having an impact on electrical, mechanical and thermal properties of the conductive tracks screen-printed on magnetic sheets is a type of paste used. The samples made with the paste PM-406 exhibited lower resistance because of a higher layer thickness compared to the lines printed with the paste PF-050. The PM-406 layers were revealed to be less durable to mechanical and thermal exposures. An analogical relationship was noticed for the samples made with PM-406 and PF-050 on a PEN foil after thermal shocks and cyclic bending. When magnetic sheets were used as a substrate, a bigger degree of damage was observed for the PF-050 samples, which even lost their electrical continuity after 1,000 bending cycles and thermal cycles, irrespective of their number. Some damage was also noticed in the magnetic sheet after the bending and thermal cycles.
Research limitations/implications
Further investigations are required to examine the influence of other types of thermal exposure on electrical properties of lines printed on magnetic sheets. Other types of magnetic sheets are also recommended to be investigated as substrate materials.
Practical implications
The results reported in this study can be useful among others for designers of radio frequency identification (RFID) systems, which are intended to operate in a challenging environment with strong mechanical and thermal exposures.
Originality/value
This paper contains valuable information concerning mechanical and thermal properties of conductive tracks screen-printed on magnetic sheets which can be used, i.e. for designing of reliable near field communication/high frequency (NFC/HF)-RFID tags suitable for metallic surface.