Tech moves fast! Stay ahead of the curve with Techopedia!
Join nearly 200,000 subscribers who receive actionable tech insights from Techopedia.
Semi-supervised learning is a method used to enable machines to classify both tangible and intangible objects. The objects the machines need to classify or identify could be as varied as inferring the learning patterns of students from classroom videos to drawing inferences from data theft attempts on servers. To learn and infer about objects, machines are provided labeled, shallow information about various types of data based on which the machines need to learn from large, structured and unstructured data they receive regularly.
The little bit of labeled data provided to the systems serve as the starting point for the computer systems. After that, the systems need to accept and learn from large volumes of unlabeled data. However, the labeled data provided may be helpful in classifying the broad type of unlabeled data the system may be receiving. For example, as labeled data, temperatures greater than 104° F should be treated as a case of high fever is given, but in reality, such high temperature may also be because of other complications. It is for the systems to use the basic labeled data and learn more about the large volumes of unlabeled data it receives. Theoretically, semi-supervised learning may be considered a better training method for systems than supervised or unsupervised learning.