What is machine learning?

“Machine learning” is the term used to describe an artificial intelligence program that can improve over time without the need for direct human intervention. “In essence, machine learning is all about analyzing big data — the automatic extraction of information and using it to make predictions, decipher whether the prediction was correct, and if incorrect, learning from that to make a more correct prediction in the future” (Sinnott). It is exactly what it sounds like: machines, learning. 

Part of understanding machine learning beyond the buzzword is knowing the basic process. IBM Cloud Education breaks machine learning down into four steps.

  1. Select and prepare a set of data for training: This set should be representative of the data that the AI will encounter in real life. It needs to be randomized and checked for biases and imbalances that could affect the results. “In some cases, the training data is labeled data—‘tagged’ to call out features and classifications the model will need to identify. Other data is unlabeled, and the model will need to extract those features and assign classifications on its own” (IBM)
  2. Choose an algorithm to run on the training data set: An algorithm is a set of statistical processing steps. It is what the AI uses to sift through data, basically the path that it follows. Its “thought process,” you might say. The type of algorithm may change based on the type of data (labeled vs unlabeled) and what the desired results are. For more details on the types of algorithms, check out IBM Cloud Education’s full article.
  3. Train the algorithm to create the model: Training involves running training data through an algorithm, changing variables, and checking the algorithm’s results against what they should be. This allows programmers to refine the algorithm, so that when it is presented with new data, it can give the most accurate results. This trained, accurate algorithm is referred to as the “model.” 
  4. Use and improve the model: As the model deals with real-world data, it continues to learn. “For example, a machine learning model designed to identify spam will ingest email messages, whereas a machine learning model that drives a robot vacuum cleaner will ingest data resulting from real-world interaction with moved furniture or new objects in the room” (IBM). 

The primary aim of all of this is to eventually have machines that can learn without human intervention, allowing them to adapt to an ever-changing world just as we humans do every day. 

Expert.ai Team. “What Is Machine Learning? A Definition.Expert.ai, 6 May 2020.IBM Cloud Education. “Machine Learning.” IBM Cloud Learn Hub, 15 July, 2020.


Loading ...