Improving the Character Recognition Efficiency of Feed Forward BP Neural Network

Fan

2018

IJGI

The increase of consumer income has resulted in the rapid development of the retail industry in China, which provides high market potential for retail companies worldwide. However, site selection for retail shops has been a confusing business issue in practical business decisions. In this study, a two-step hybrid model in site selection for small retail shops was proposed. The two steps were spatial accessibility evaluation and market potential estimation. The spatial accessibility of target regions was evaluated based on the improved gravity model to determine regions that lack retail shops. Then, a PCA (principal component analysis)-BP (backpropagation network) model was established to estimate the market potential in the target regions. The two-step model could determine sites with the most market potential and low competition. We conducted the experiment in Guiyang, China and considered 18 socioeconomic factors to make the site selection convincing. Through the experiment, 42 locations were determined with high business value; the locations were recommended to the new retail shops. The accuracy of the PCA-BP model was then proven satisfactory by comparing it with other regression methods. The proposed model could guide retail chains in enhancing business location planning and formulating regional development policies.

Section: Accuracy Analysis Of the Modelmentioning

confidence: 99%

Site Selection of Retail Shops Based on Spatial Accessibility and Hybrid BP Neural Network

Fan

2018

IJGI

“…All neurons are organized into layers; the sequence of layers defines the order in which the activations are computed [8]. Back propagation is the best known and widely used learning algorithm in training multi-layers, such as the Feed Forward Neural Networks [9]. The architecture of the Feed Forward Back-propagation Neural Network is presented in [10].…”

Section: Application Of Neural Network In Optimization Functionmentioning

confidence: 99%

Combining FEM and neural networking in the design of optimization of traditional Montenegrin chair

Vujović¹,

Krivokapic²,

Grujicic³

et al. 2016

FME Transaction

The paper presents the results of tests and analysis of strength of traditional Montenegrin chair using FEM (The Finite Element Method) according to the procedure described in the relevant European standards. Stresses in the critical points of the structure were obtained and compared with values of permissible stresses for appropriate material. The influence of design parameters of chair on distribution and intensity of the stresses was considered, so it was selected the appropriate combination of parameters that presents the optimal solution from the aspect of the mentioned criterion. The obtained solution is compared with a volume equivalent model of usual chair with four legs. Based on the results we trained neural network in order to make it enable selection of design parameters of the traditional Montenegrin chairs that provide a satisfactory solution

“…Then the network was trained and momentum term was used to modify the weights at each epoch. [2]. Another technique follows use of the training patterns in some random order such that the weights are adjusted based on the back propagation algorithms.…”

Section: Review Of Literaturementioning

confidence: 99%

Designing an Intelligent System for Optical Handwritten Character Recognition using ANN

Mahajan¹,

Mahajan²

2014

IJCA

OCR is a technology which is unique in itself and is used for the various applications involving scanning and recognition. Although recognition of the slant and cursivity of a handwritten text is the major field of research but the recognition rate of the hand written text is extremely lower than the hand printed text. This is because for the recognition of hand written text use of the contextual information and grammar associated with the grammar is required i.e. whether the word identified is a verb, a noun or an adjective. Thus with the recognition of the characters we need to recognize a word completely and along with that find the respective properties of the word according to semantics of the sentence. This helps to develop an intelligent system for the recognition of the handwriting of a person. In this paper, an intelligent system for "OPTICAL CHARACTER RECOGINITION" using Artificial Neural Network based approach and a Feature Extraction algorithm before an ANN can be applied for classification of characters for the character recognition is proposed.