You’ve likely heard the term “machine learning” recently. Although the concept has been around for awhile, it is now being put to practical use in a number of applications. Applications getting the most attention are those that solve “big” problems, such as world hunger or easing traffic congestion. IBM Watson is perhaps the best-known machine learning system.
Machine learning is powered by software that allows a computer to learn by analyzing patterns in data and outcomes. Over time and without additional programming, the machine learning system is able to predict outcomes based on previous events. It can then recommend or automatically execute an action. Some retailers, for example, are using machine learning to help optimize inventory levels or even adjust pricing.
Traditional statistical analysis relies on human assumptions of what the data structure looks like. Machine learning allows the computer to identify the data structure through an iterative process, testing and revising as it goes. As a result, a machine learning system can learn the structure of a dataset much faster than a human and consequently apply that understanding to achieving a specific result.
A common misconception about machine learning is that you have to be a big company to need or afford it. There is a growing base of applications and development tools, many of them cloud-based, that offer affordable machine-learning capability, off-the-shelf or do-it-yourself. Some of them are useful to the digital imaging community and its customers.
For example, most companies are always looking for ways to get more from their sales and marketing efforts. Sales and marketing are very data and statistics driven as companies try to identify and qualify prospects from a pool of contacts. A number of applications use machine learning to accelerate that process. Let’s quickly look at one of them, predictive intelligence platform 6Sense.
6Sense is targeted to B2B marketers who want to identify prospects who are close to making a buying decision. It does so by analyzing both structured (like that stored in a database) and unstructured data (text from emails, scanned documents, etc.) to generate a model for customers’ intent to buy. This allows sales teams to qualify prospects more accurately so they focus their efforts more efficiently.
Canopy Labs is a cloud-based marketing tool that employs machine learning techniques to gain insight from customers’ behavior across multiple marketing channels. Its simple, tiered pricing structure makes it accessible to even the smallest businesses.
Most high-end enterprise analytics tools such as SAS and SAP are incorporating machine learning capabilities that can be customized to any application. That option is available to companies not yet at the enterprise level. IBM Watson, for example, has cloud-based editions priced well within the means of most small businesses. And despite its name, BigML is a machine-language decision support tool aimed at small- to medium-sized businesses.
For dealers offering managed services or software solutions, adding machine language tools to your offerings could be a great value-add. With those tools, you could build much richer reporting systems or automate decision making within key business processes. This would give clients greater insight into their business and save both time and money. Those same benefits are within reach of any digital imaging business as well.
