Machine Learning, what ist this?
Machine learning is a generic term for the “artificial” generation of knowledge from experience. In order for software to learn independently and also to find solutions, people have to act in advance. You must provide the systems with the data and algorithms necessary for machine learning. First of all, however, the rules for analyzing the database and recognizing patterns must be defined. This enables systems to perform the following tasks:
- Process optimization based on recognized patterns.
- independent adaptation to developments.
- Make predictions based on analyzed data.
- Calculating probabilities of certain events.
- Find, summarize and extract relevant data.
Different types of Machine Learning
Algorithms play a central role in machine learning because they are responsible for recognizing patterns and generating suitable solutions. Algorithms can be divided into five different categories:
- Supervised learning: Here, sample models are defined and specified in advance in order to assign information appropriately to the model groups of the algorithms. The goal is to train the system in the context of successive arithmetic operations with different inputs and outputs.
- Unsupervised learning: here the model groups are automated based on independently recognized patterns. The machine tries to structure and sort the entered data according to certain characteristics.
- Partially supervised learning: here is a mixture of both methods.
- Empowering Learning: This type of learning is based on rewards and punishments. This interaction tells the algorithm how to respond to different situations.
- Active learning: Here the algorithm offers the possibility to request the desired results for certain input data. To minimize the number of questions, the algorithm itself selects relevant questions with high relevance to the results.
Next page