Predictive modeling is a process that uses data mining and probability to forecast outcomes. Each model is made up of a number of predictors, which are variables that are likely to influence future results. Once data has been collected for relevant predictors, a statistical model is formulated. As additional data becomes available, the statistical analysis model is validated or revised.
Predictive modeling can and has been used in a variety of applications. Most often it is associated with meteorology and weather forecasting, but it has many applications in business as well.
Some of the most common uses of predictive modeling touches everyone at some point, whether you are aware of it or not. Online advertising and marketing professions use a web surfers’ historical data, running it through algorithms to determine what kinds of products users might be interested in and what they are likely to click on. In customer relationship management, predictive modeling is used to target messaging to customers who are most likely to make a purchase. Think of the familiar, “shoppers who bought this also were interested in this” product pitch on many website, but probably the most familiar is Amazon.
Another use is to identify whether email messages are spam or legitimate. Bayesian spam filters use predictive modeling to identify the probability that a given message is spam. In fraud detection, predictive modeling is used to identify outliers in a data set that point toward fraudulent activity.
Those are the common uses, but the most complex area of predictive modeling is the neural network. This type of machine learning model independently reviews large volumes of labeled data in search of correlations between variables in the data.
It can detect even subtle correlations that only emerge after reviewing millions of data points. The algorithm can then make inferences about unlabeled data files that are similar in type to the data set it trained on. All of this is done in seconds vs. the impossible amount of time it would take humans.
Neural networks form the basis of many of today’s examples of artificial intelligence (AI), including image recognition, smart assistants and natural language generation (NLG).
Predictive modeling is often considered to be primarily a mathematical problem. However, users must plan for the technical and organizational barriers that might prevent them from getting the data they need. For example, systems that store useful data are not always connected directly to centralized data warehouses.
Predictive models often perform calculations during live transactions—for example, to evaluate the risk or opportunity of a given customer or transaction to guide a decision. For example, if health insurers could accurately predict secular trends, premiums would be set appropriately, profit targets would be met with more consistency, and health insurers would be more competitive in the marketplace.
Melody K. Smith
Sponsored by Access Innovations, the world leader in thesaurus, ontology, and taxonomy creation and metadata application.
