Since the beginning of the new millenium, we have witnessed neural networks come of age. The idea of learning to solve complex pattern recognition problems using an intelligent data-driven approach is no longer simply an interesting challenge for academic researchers. Neural networks have proven themselves to be a valuable tool across a wide range of functional areas affecting most businesses. As a critical component of most data mining systems, they are also changing the way organizations view the relationship between their data and their business strategy.

Neural networks are simple computational tools for examining data and developing models that help to identify interesting patterns or structures in the data. The data used to develop these models is known as training data. Once a neural network has been exposed to the training data, and has learned the patterns that exist in that data, it can be applied to new data thereby achieving a variety of outcomes. Neural networks can be used to

While there are many different neural network models that have been developed over the last fifty years or so to achieve these tasks of prediction, classification, and clustering, we will be focusing only on one of the main model that  have successfully found application across a broad range of business areas. We call this model a multilayered feedforward neural network (MFNN) and is an example of a neural network trained with supervised learning (Rumelhart & McClelland, 1986).

With supervised learning models, the training data contains complete information about the characteristics of the data and the observable outcomes. Models can be developed that learn the relationship between these characteristics (inputs) and outcomes (outputs). Using a MFNN to model the relationship between money spent during last week’s advertising campaign and this week’s sales figures is an example of a prediction application. An example of a related classification application is using a MFNN to model the relationship between a customer’s demographic characteristics and their status as a high-value or low-value customer. For both of these example applications, the training data must contain numeric information on both the inputs and the outputs in order for the MFNN to generate a model. The MFNN is repeatedly trained with this data until it learns to represent these relationships correctly.

For a given input pattern, the network produces an output (or set of outputs) zk, and this response is compared to the known desired response of each neuron dk. For classification problems, the desired response of each neuron will be either zero or one, while for prediction problems it tends to be continuous valued. The weights of the network are then modified to correct or reduce the error, and the next pattern is presented. The weights are continually modified in this manner until the total error across all training patterns is reduced below some pre-defined tolerance level. This learning algorithm is known as the backpropagation algorithm (Werbos, 1974; Le Cun, 1985; Parker, 1985). Proof that the effect of these weight updates gradually minimizes the mean squared error (MSE) across all input patterns relies on the fact that the backpropagation learning algorithm performs gradient descent on the error function.

The main steps of the backpropagation learning algorithm are summarized below:

Step 1: Input training vector.
Step 2: Hidden nodes calculate their outputs.
Step 3: Output nodes calculate their outputs on the basis of Step 2.
Step 4: Calculate the differences between the results of Step 3 and targets.
Step 5: Apply the first part of the training rule using the results of Step 4.
Step 6: For each hidden node, n, calculate d(n).
Step 7: Apply the second part of the training rule using the results of Step 6.

Steps 1 through 3 are often called the forward pass, and steps 4 through 7 are often called the backward pass. Hence, the name: back-propagation.

Real life applications

The tasks to which artificial neural networks are applied tend to fall within the following broad categories:

• Function approximation, or regression analysis, including time series prediction and modelling.
• Classification, including pattern and sequence recognition, novelty detection and sequential decision making.
• Data processing, including filtering, clustering and compression.

Application areas include and not restricted to

• system identification and control (vehicle control, process control),
• game-playing and decision making (backgammon, chess, racing),
• pattern recognition (radar systems, face identification, object recognition and more),
• sequence recognition (gesture, speech, handwritten text recognition),
• medical diagnosis, financial applications,
• data mining (or knowledge discovery in databases, "KDD"),
• visualisation and
• email spam filtering.

We will examine some detailed case studies demonstrating how the MFNN (4Cast XL) has been successfully applied to problems as diverse as sales forecasting, direct marketing, credit scoring, exchange rate modelling, business strategy evaluation, insurance claim analysis, and assessing the value of customers.

Neural networks have been applied to numerous situations where time series prediction is required – predicting weather, climate, stocks and share prices, currency exchange rates, airline passengers, etc. We can turn the temporal problem into a simple inputoutput mapping by taking the time series data x(t) at k time-slices t, t–1, t–2, …, t–k+1 as the inputs, and the output is the prediction for x(t+1).