scholarly journals Influence of Training Set Selection in Artificial Neural Network-Based Propagation Path Loss Predictions

2012 ◽  
Vol 2012 ◽  
pp. 1-7 ◽  
Author(s):  
Ignacio Fernández Anitzine ◽  
Juan Antonio Romo Argota ◽  
Fernado Pérez Fontán
Keyword(s):  
Path Loss ◽  
Propagation Path ◽  
Training Set ◽  
Received Power ◽  
Optimum Selection ◽  
Combined Use ◽  
Artificial Neural ◽  
Training Set Selection ◽  
And Training ◽  
Selection Of

This paper analyzes the use of artificial neural networks (ANNs) for predicting the received power/path loss in both outdoor and indoor links. The approach followed has been a combined use of ANNs and ray-tracing, the latter allowing the identification and parameterization of the so-called dominant path. A complete description of the process for creating and training an ANN-based model is presented with special emphasis on the training process. More specifically, we will be discussing various techniques to arrive at valid predictions focusing on an optimum selection of the training set. A quantitative analysis based on results from two narrowband measurement campaigns, one outdoors and the other indoors, is also presented.

Evaluating the Use of Artificial Neural Networks, Graph Theory, and Complexity Theory to Predict Automotive Assembly Defects

2016 ◽  
Author(s):  
Apurva Patel ◽  
Patrick Andrews ◽  
Joshua D. Summers
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Prediction Models ◽  
Assembly Model ◽  
Training Set ◽  
Automotive Assembly ◽  
Artificial Neural ◽  
Training Set Selection ◽  
Selection Of ◽  
Assembly Models

Artificial Neural Networks (ANNs) have been used to predict assembly time and market value from assembly models. This was done by converting the assembly models into bipartite graphs and extracting 29 graph complexity metrics which were used to train the ANN prediction models. This paper presents the use of sub-assembly models instead of the entire assembly model to predict assembly quality defects at an automotive OEM. The size of the training set, order of the bipartite graph, selection of training set, and defect type were experimentally studied. With a training size of 28 parts, an interpolation focused training set selection, and second order graph seeding, over 70% of the predictions were within 100% of the target value. The study shows that with an increase in training size and careful selection of training sets, assembly defects can be predicted reliably from sub-assemblies complexity data.

Evaluating the Use of Artificial Neural Networks and Graph Complexity to Predict Automotive Assembly Quality Defects

2017 ◽  
Vol 17 (3) ◽  
Author(s):  
Apurva Patel ◽  
Patrick Andrews ◽  
Joshua D. Summers ◽  
Erin Harrison ◽  
Joerg Schulte ◽  
...  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Prediction Models ◽  
Assembly Model ◽  
Training Set ◽  
Assembly Quality ◽  
Automotive Assembly ◽  
Artificial Neural ◽  
Training Set Selection ◽  
Selection Of

This paper presents the use of subassembly models instead of the entire assembly model to predict assembly quality defects at an automotive original equipment manufacturer (OEM). Specifically, artificial neural networks (ANNs) were used to predict assembly time and market value from assembly models. These models were converted into bipartite graphs from which 29 graph complexity metrics were extracted to train 18,900 ANN prediction models. The size of the training set, order of the bipartite graph, selection of training set, and defect type were experimentally studied. With a training size of 28 parts, an interpolation focused training set selection with a second-order graph seeding ensured that 70% of all predictions were within 100% of the target value. The study shows that with an increase in training size and careful selection of training sets, assembly defects can be predicted reliably from subassemblies' complexity data.

scholarly journals Determination of an Outdoor Path Loss Model and Signal Penetration Level in Some Selected Modern Residential and Office Apartments in Ogbomosho, Oyo State, Nigeria

2018 ◽  
pp. 1-25
Author(s):  
V. O. A. Akpaida ◽  
F. I. Anyasi ◽  
S. I. Uzairue ◽  
A. I. Idim
Keyword(s):  
Building Materials ◽  
Path Loss ◽  
Signal Strength ◽  
Base Stations ◽  
Propagation Path ◽  
Loss Model ◽  
Path Loss Model ◽  
Path Loss Exponent ◽  
Received Power

This article involves the site specific determination of an outdoor path loss model and Signal penetration level in some selected modern residential and office apartments in Ogbomosho, Oyo State. Measurements of signal strength and its associated location parameters referenced globally were carried out. Propagation path loss characteristics of Ogbomosho were investigated using three different locations with distinctively different yet modern building materials. Consequently, received signal strength (RSS) was measured at a distance d in meters, from appropriate base stations for various environments investigated. The data were analyzed to determine the propagation path loss exponent, signal penetration level and path loss characteristics. From calculations, the average building penetration losses were, 5.93dBm, 6.40dBm and 6.1dBm outside the hollow blocks B1, solid blocks B2 and hollow blocks mixed with pre cast asbestos B3, buildings respectively with a corresponding path loss exponent values of, 3.77, 3.80 and 3.63. Models were developed and validated, and used to predict the received power inside specific buildings. Moreover, the propagation models developed for the different building types can be used to predict the respective signal level within the building types, once the transmitter – receiver distance is known. The readings obtained from the developed models were compared with both the measured values and values computed using some existing models with satisfactory results obtained.

The importance of outlier detection and training set selection for reliable environmental QSAR predictions

Chemosphere ◽  
2006 ◽  
Vol 63 (1) ◽  
pp. 99-108 ◽  
Author(s):  
Erik Furusjö ◽  
Anders Svenson ◽  
Magnus Rahmberg ◽  
Magnus Andersson
Keyword(s):  
Outlier Detection ◽  
Training Set ◽  
Selection For ◽  
Training Set Selection ◽  
And Training

Computational methods for training set selection and error assessment applied to catalyst design: guidelines for deciding which reactions to run first and which to run next

2021 ◽  
Author(s):  
Andrew F. Zahrt ◽  
Brennan T. Rose ◽  
William T. Darrow ◽  
Jeremy J. Henle ◽  
Scott E. Denmark
Keyword(s):  
In Silico ◽  
Subset Selection ◽  
Design Guidelines ◽  
Catalyst Design ◽  
Error Assessment ◽  
Selection Methods ◽  
Training Set ◽  
Training Set Selection ◽  
Catalyst Selection ◽  
Selection Of

Different subset selection methods are examined to guide catalyst selection in optimization campaigns. Error assessment methods are used to quantitatively inform selection of new catalyst candidates from in silico libraries of catalyst structures.

Sign in / Sign up

Export Citation Format

Share Document