Chapter 6. Real-Time Machine Learning Applications
Combining terms like “real time” and “machine learning” runs the risk of drifting into the realm of buzz and away from business problems. However, improvements in real-time data processing systems and the ready availability of machine learning libraries make it so that applying machine learning to real-time problems is not only possible, but in many cases simply requires connecting the dots on a few crucial components.
The stereotype about data science is that its practitioners operate in silos, issuing declarations about data from on high, removed from the operational aspects of a business. This mindset reflects the latency and difficulty associated with legacy data analysis and processing toolchains. In contrast, modern database and data processing system design must embrace modularity and accessibility of data.
Real-Time Applications of Supervised Learning
The power of supervised learning applies to numerous business problems. Regression is familiar to any data scientist, finance or risk analyst, or anyone who took a statistics class in college. What has changed recently is the availability of powerful data processing software that enables businesses to apply these tools to extremely low-latency problems.
Any system that automates data-driven decision making, for instance, will need to not only build and train a model, but use the model to score or make predictions. When developing a model, data scientists ...