Network structure determination is an important issue in pattern classification based on a probabilistic neural network. In this study, a supervised network structure determination algorithm is proposed. The proposed algorithm consists of two parts and runs in an iterative way. The first part identifies an appropriate smoothing parameter using a genetic algorithm, while the second part determines suitable pattern layer neurons using a forward regression orthogonal algorithm. The proposed algorithm is capable of offering a fairly small network structure with satisfactory classification accuracy.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/72.857781 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Characteristic Structure of the Relationships Among Depression, Maladaptive Cognition, and Internet Addiction and Its Temporal Evolution in University Students: A Network Analysis.
Stress Health
February 2025
Collaborative Innovation Center of Assessment Toward Basic Education Quality, Beijing Normal University, Beijing, China.
This study explored the structure and temporal evolution of the relationship among depression, maladaptive cognition, and internet addiction (DMI) among university students by focusing on topological and dynamic properties in a network analysis. A 3-year longitudinal survey was conducted with 873 university students (M = 18.32, SD = 0.
View Article and Find Full Text PDFThe Digital Health Competencies in Medical Education Framework: An International Consensus Statement Based on a Delphi Study.
JAMA Netw Open
January 2025
Faculty of Medicine, Sana'a University, Sana'a, Yemen.
Importance: Rapid digitalization of health care and a dearth of digital health education for medical students and junior physicians worldwide means there is an imperative for more training in this dynamic and evolving field.
Objective: To develop an evidence-informed, consensus-guided, adaptable digital health competencies framework for the design and development of digital health curricula in medical institutions globally.
Evidence Review: A core group was assembled to oversee the development of the Digital Health Competencies in Medical Education (DECODE) framework.
What Predicts Early Math in Autism? A Study of Cognitive and Linguistic Factors.
J Autism Dev Disord
January 2025
Department of Linguistics and Basque Studies, Universidad del País Vasco (UPV/EHU), Paseo de la Universidad 5, Vitoria-Gasteiz, 01006, Spain.
This study aimed to examine early mathematical abilities in young children with autism aged four to seven without intellectual disabilities and their connection with autism severity, non-verbal intelligence, and linguistic abilities (receptive vocabulary and grammar). The study involved 42 children with autism. We assessed participants' cognitive, mathematical, and linguistic abilities.
View Article and Find Full Text PDFTechnological and Functional Potentialities of Mairá (Casimirella rupestris) and Ariá (Goeppertia allouia) Starches.
Plant Foods Hum Nutr
January 2025
Food4You Research Group, Department of Food Science and Nutrition (DECAN), Faculty of Food Engineering (FEA), Universidade Estadual de Campinas (UNICAMP), Campinas, SP, Brazil.
The Amazon region holds untapped potential with its starch-rich tubers, which are not yet industrialized and face a risk of extinction due to competition from widely cultivated crops. Beyond their traditional subsistence use, Amazonian tubers such as Mairá and Ariá can be utilized as starch sources, offering an opportunity to support regional agriculture, preserve indigenous heritage, and provide sustainable income streams. This study aimed to characterize starches extracted from Mairá (MPS) and Ariá (ARS for rhizome and APS for potato), focusing on their technological and functional potential.
View Article and Find Full Text PDFCocatalysis of catalytic hairpin assembly and ternary heterojunction BiS@MoS@BiMoO promotes ultra-sensitive electrochemical of detection MiRNA-21.
Mikrochim Acta
January 2025
USST-UH International Joint Laboratory for Tumor Diagnosis and Energy Treatment, University of Shanghai for Science and Technology, Shanghai, 200093, China.
Ternary heterojunction BiS/MoS/BiMoO was designed as a signal probe to develop a dual signal amplification strategy empowered electrochemical biosensor for sensitive miRNA-21 detection by combining with catalytic hairpin assembly (CHA). The combination of the BiS/MoS/BiMoO heterojunction as a tracer indication probe and the CHA amplification strategy not only took fully use of the highly dense nanowire interwoven structure and superior active region of the probe, but also endowed the ability to improve the molecular hybridization efficiency by collision, which significantly avoided the cumbersome chain design and greatly simplified the step-by-step construction of the electrode surface. Hairpin H1 was first added dropwise to the gold nanoparticle-decorated electrode surface, and then opened by the introduced miRNA-21 to initiate the specific hybridization.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!