Machine learning (ML) is a field of computer science that gives computers the ability to learn without being explicitly programmed. In BI, this general definition is expanded to include all automatic tasks and programs built to find new patterns and information buried in your data. It therefore includes tools for 'heuristics', automated 'data mining', 'predictive analysis', 'deep learning' and ultimately 'AI' in the context of data and its analysis.
- Click here to review a listing on Machine Learning algorithms supported by Pyramid.
In Pyramid, machine learning is manifested in a variety of ways. In fact, much of the platform is designed around the idea that ML is and will continue to be central to the BI platform and its users.
Key examples of where Pyramid uses ML as a tool itself include:
ML is used to drive the Smart Discovery tool-set, which automatically analyses your models, finds interesting relationships and patterns in that model and automatically builds the relevant reports and dashboards to highlight these findings.
- ML is used to find the most relevant reports in the repository for specific users based on the history of other users and their patterns
- ML is used to drive a variety of heuristic engines in the application to guess the best model designs; suggest the next click suggestion in discovery; choose the best data forecasting algorithm (itself ML logic); find data outliers; auto-suggest how columns should be classified; and many more.
Pyramid also facilitates the usage of ML by its users. Indeed, this is the most critical ML capability: a platform for users to deploy ML functions and libraries against their own data sets to auto-discover information, make predictions, find patterns and ultimately make better decisions. Key examples include:
- The various ML libraries available for use through simple point-and-click prompts in Modeling including functions from the Weka, MLIB and TensorFlow libraries, scriptlets using R and Python, as well as Pyramid-built functions specifically aimed at classic ML requirements in BI.
- The ML marketplace complete with a variety of typical ML functions presented as predefined scripts
- Various high powered scripting and programming engines that include vast libraries of ML related logic. Specifically R, Python and base SAS.
- The ML scripting tools for creating shared ML scripts within an organization in Formulations