Machine learning is a branch of AI (artificial intelligence) that consists of deep learning, supervised and unsupervised learning. It is the convergence of computer science and mathematical branches of statistics and calculus. It can be described as the automatic application of mathematical techniques on a massive scale using volumes of data enabled by extremely fast computing speeds. Some circles consider this to be machines imitating intelligent behaviour.
With the availability of large volumes of data and extremely fast computing capability together with easy access to latest digital techniques using open source libraries, digital approaches to plant optimisation and orebody understanding using AI and machine learning are becoming commonplace. However, there is a persistent perception in industry that mining data is lacking and of low quality therefore will generate ineffective and bad prediction models.
The concerns include insufficient data (missing data), data of poor quality (erroneous fields and strings) or data that is just wrong (values are not accurate). This is compounded by the fact that data is located and stored in many disparate systems and databases.