Machine Learning AI

What is Machine Learning?

Machine Learning (ML) is a cutting edge form of AI. It's the type of AI that is generally credited with driving your Tesla down the highway and promises to yield the biggest impact for everyday workplaces one day.

Machine Learning is about finding patterns in big data where commonplace statistical analysis doesn’t see any, and taking such pattern to predict results without much human interpretation.

ML however requires three key ingredients to become effective:

1. ML requires lots of data

To teach the AI new tricks it requires bucket loads of data into its model input to make reliable output scoring. Tesla for example has deployed an auto steering feature to its cars which simultaneously sends home all the data points it collects, interventions by the driver, successful evasions, false alarms, etc. to learn from the mistakes and gradually sharpen the senses. A great way to produce a lot of input is through sensors: Either your hardware has built-in ones like radar, cameras, steering wheel, etc. (if it's a car) – or you lean on IoT (Internet of Things). Bluetooth beacons, health trackers, smart home sensors, public databases, etc. are just a small fraction of the ever growing number of internet-connected sensors that can generate much data (too much for any normal human to process)

2. ML needs good discovery

To make sense of your data and cut through the noise Machine Learning puts algorithms at work that can sort, slice and translate a data chaos into comprehensible insights. (If you want to weird out your colleagues listen to sound of different sorting algorithms at work:

There are two ways for algorithms to learn about the data, unsupervised or supervised.

  • Unsupervised ones deal with figures and raw data only, so there are no descriptive labels or dependant variables established. The aim for the algorithm is to find an intrinsic structure where humans didn’t think there would be one. This is useful for gaining new insights into market segmentation, correlations, outliers, etc.
  • On the other hand, supervised algorithms have knowledge about relations between different data sets through labels and variables and use their power primarily to extrapolate and predict future data. This might come in handy for anything from climate change models, predictive analytics, content recommendations, etc.

3. Machine Learning Deployment challenges

Machine Learning needs to find its way from the computer science labs into softwareMore vendors like CRM, Marketing, ERP, etc. are building up competencies in either embedding Machine Learning or integrating tightly with services that offer it.