By Dean Whittaker

While attending a geek conference, Big Data Ignite, this month I purchased a copy of “The Book of Why.” In it the authors, Judea Pear and Dana Mackenzie, discuss the science of cause and effect which is based on mathematics. As with any science, it endeavors to predict the outcome from data using tools such as predictive analytics and machine learning, both of which fall under the umbrella of artificial intelligence.

Predictive analytics encompasses a variety of statistical techniques from data mining, predictive modelling, and machine learning, that analyze current and historical facts to make predictions about future or otherwise unknown events. Machine learning is a field of computer science that uses statistical techniques to give computer systems the ability to “learn” with data, without being explicitly programmed. (Wikipedia)

One of the most in demand occupations is that of data scientists as companies seek insights from data they collect mostly about us and about their business. Big Data Ignite conference participants included data analysts from office furniture manufactures, drug companies, and apparel manufactures, all looking for ways to improve their manufacturing processes and better understand their customers. Drug companies use machine learning to predict which disease-fighting molecules are the most effective. Food companies use data analytics to understand the patterns in our purchases to determine what foods to put in the supply chain. Office furniture manufacturers use sensing data to show available desks, meeting, and conference rooms in an office building using a phone app.

Applying both data analytics and machine learning to economic development is my intent. In the past, we have collected data on new and expand announcements for several years, and more recently, we have been collecting data on business layoffs and closures. We look for patterns and trends to predict companies likely to relocate or expand in the next 12-18 months. We have validated our predictive analytic algorithm (a weighted matrix formula) by comparing our predictions to actual new and expanded company data. The encouraging result is that our current predictive analytic process is approximately 34% accurate. Because of machine learning self-programing capability at pattern recognition, we hope to significantly enhance the accuracy of our predictions. Stay tuned.