Prerequisites for Machine Learning
Prerequisites for machine learning are essentially three key elements followed by certain skills in order to get the most out of learning from big data. The first prerequisite is that there must be a pattern in the data to look for. If such a pattern not exist then there is hardly anything machine learning can learn from the data. The best known example in this context is a random number generator that generated randomly data items. Learning a pattern in this random data items is not useful.
Another element of the prerequisites for machine learning is typically that there is no existing formula for the problem at hand already. One example is numerical weather prediction that uses known physical laws and numerical formulas in order to determine the weather for the next days. Given that there is such a formula is makes hardly sense to predict the weather based solely on machine learning.
The third key element in the prerequisites for machine learning is to have enough data available to learn from. Machine learning is learning from data. Although it does not need to be always big data there must be a certain amount of data available before learning actually makes sense.
The above three key elements are important but another necessary part of the prerequisites for machine learning are the right skills shown in the image above. In order to perform data science that is a term often used in the context of learning from data one must have a certain level of skills in the illustrated fields. The idea ‘learning from data‘ is shared with a wide variety of other disciplines like in signal processing, data mining, statistics, and many others with a huge overlap.
Prerequisites for Machine Learning Details
The following video is a good source to watch in this context:
