Business Applications Of Neural Networks: The State-of-the-art Of Real-world ApplicationsBill Edisbury, Paulo J G Lisboa, Alfredo Vellido World Scientific, 2000. gada 30. aug. - 220 lappuses Neural networks are increasingly being used in real-world business applications and, in some cases, such as fraud detection, they have already become the method of choice. Their use for risk assessment is also growing and they have been employed to visualise complex databases for marketing segmentation. This boom in applications covers a wide range of business interests — from finance management, through forecasting, to production. The combination of statistical, neural and fuzzy methods now enables direct quantitative studies to be carried out without the need for rocket-science expertise.This book reviews the state-of-the-art in current applications of neural-network methods in three important areas of business analysis. It includes a tutorial chapter to introduce new users to the potential and pitfalls of this new technology. |
No grāmatas satura
1.–5. rezultāts no 85.
ix. lappuse
... figure 1. The saturation curve in figure 2 represents the neuron's firing rate, which reduces under inhibition, or negative activation, and increases under positive excitation up to a maximum firing frequency, denoted here as a unity ...
... figure 1. The saturation curve in figure 2 represents the neuron's firing rate, which reduces under inhibition, or negative activation, and increases under positive excitation up to a maximum firing frequency, denoted here as a unity ...
x. lappuse
... Figure 2: Typical response function for a single neuron. This curve has the functional form y = 1/[1+exp(-a)]. The illustrative data for this simplified credit risk example, for the input into a single linear cell, are shown in figure 3 ...
... Figure 2: Typical response function for a single neuron. This curve has the functional form y = 1/[1+exp(-a)]. The illustrative data for this simplified credit risk example, for the input into a single linear cell, are shown in figure 3 ...
xi. lappuse
... Figure 3: Example of a single neural cell, in this example exhibiting a linear response to the network activation. If the synaptic strengths assume the values shown, this cell will correctly recall four data patterns which represent ...
... Figure 3: Example of a single neural cell, in this example exhibiting a linear response to the network activation. If the synaptic strengths assume the values shown, this cell will correctly recall four data patterns which represent ...
xii. lappuse
... figure 5. The cell layer sandwiched between the input nodes on the left, and the response nodes on the right, in figure 5, is called a hidden layer. The network weights are estimated using variants of gradient descent algorithms, taking ...
... figure 5. The cell layer sandwiched between the input nodes on the left, and the response nodes on the right, in figure 5, is called a hidden layer. The network weights are estimated using variants of gradient descent algorithms, taking ...
xiii. lappuse
Bill Edisbury, Paulo J G Lisboa, Alfredo Vellido. Figure 5: A Multi-layer Perceptron (MLP) neural network. The network inputs are labelled x, the hidden nodes by h and the predicted output by y. The remaining symbols indicate connection ...
Bill Edisbury, Paulo J G Lisboa, Alfredo Vellido. Figure 5: A Multi-layer Perceptron (MLP) neural network. The network inputs are labelled x, the hidden nodes by h and the predicted output by y. The remaining symbols indicate connection ...
Saturs
1 | |
Chapter 2 Extracting Rules Concerning Market Segmentation from Artificial Neural Networks | 13 |
Chapter 3 Characterising and Segmenting the BusinesstoConsumer ECommerce Market Using Neural Networks | 29 |
Chapter 4 A Neurofuzzy Model for Predicting Business Bankruptcy | 55 |
Chapter 5 Neural Networks for Analysis of Financial Statements | 73 |
Chapter 6 Developments in Accurate Consumer Risk Assessment Technology | 85 |
Chapter 7 Strategies for Exploiting Neural Networks in Retail Finance | 99 |
Chapter 8 Novel Techniques for Profiling and Fraud Detection in Mobile Telecommunications | 113 |
Chapter 9 Detecting Payment Card Fraud with Neural Networks | 141 |
Chapter 10 Money Laundering Detection with a NeuralNetwork | 159 |
Chapter 11 Utilising Fuzzy Logic and Neurofuzzy for Business Advantage | 173 |
Index | 195 |
Citi izdevumi - Skatīt visu
Business Applications of Neural Networks: The State-of-the-art of Real-world ... Paulo J. G. Lisboa,Bill Edisbury,Alfredo Vellido Ierobežota priekšskatīšana - 2000 |
Business Applications of Neural Networks: The State-of-the-art of Real-world ... Bill Edisbury Ierobežota priekšskatīšana - 2000 |
Business Applications of Neural Networks: The State-of-the-art of Real-world ... Paulo J. G. Lisboa,Bill Edisbury,Alfredo Vellido Priekšskatījums nav pieejams - 2000 |
Bieži izmantoti vārdi un frāzes
accuracy AFPR aircraft application Artificial Neural Networks behaviour British Airways card fraud cardholder classification clusters data mining data protection data set data space database decision engine described developed example extracted factors Falcon score Figure financial ratios fraud detection system fraudster fraudulent fuzzy logic fuzzy logic system fuzzyTECH Genetic Algorithms hidden unit activation identified input variables issuers layer linear linear discriminant analysis logistic regression Machine Learning methods mobile money laundering neural network model neurofuzzy nodes non-linear optimisation output parameters patterns performance perimetry personal data predictive model problem profiles prototype pruning pruning algorithm ratios rule block samples scorecard segmentation self-organising map Self-Organizing Map Shopping experience solution statistical subset techniques telecommunications threshold Toll Tickets tool transactions unsupervised learning variable selection vector visualization weights