Version 10 introduces a wide range of integrated machine learning capabilities, from highly automated functions like Predict and Classify to functions based on specific methods and diagnostics. The functions work on many types of data—including numerical, categorical, textual, and image—allowing everyone to perform state-of-the-art machine learning in a simple way. A wide range of tasks can be performed, such as text classification, image recognition, or classification from generic data.
Version 10 includes a range of pre-trained models for a wide variety of machine learning tasks. These include models for text-based tasks that arise in data science, such as email spam detection, language detection, or social media topic identification, as well as image-based models, such as flag identification.
Version 10 now includes fully automated fitting and diagnostics across the full suite of time series processes, making time series modeling an everyday exploratory tool. Time series modeling has also been deepened to include ARCH and GARCH processes, as well as vector-valued versions of standard time series models. The full time series model framework has been greatly enhanced, including simulation, estimation, and property computations.