matomo-org/matomo

Use machine learning to create better insights

Open

#5,959 opened on Aug 9, 2014

View on GitHub
 (8 comments) (2 reactions) (0 assignees)PHP (21,513 stars) (2,847 forks)batch import
EnhancementHelp wanted

Description

Piwik makes it easy to collect data from many systems, databases, services.

All this data could be analysed to extract insights. Anyone could build plugins for the open analytics platform Piwik. A plugin can contain logic to extra insights from existing data or new custom data.

What is machine learning?

Machine learning is the science of getting computers to act without being explicitly programmed. In the past decade, machine learning has given us self-driving cars, practical speech recognition, effective web search, and a vastly improved understanding of the human genome. Machine learning is so pervasive today that you probably use it dozens of times a day without knowing it.

Source: coursera stanford machine learning course

Possibilities of machine learning algorithms for Piwik

  • Personalisation of content on the website
  • Recommendation service eg. user who bought X also bought Y and Z
  • Better data insights
    • To learn more about the basic work done so far in Piwik see #5477 and discuss it further in #5246.

Machine learning tools

  • It is so exciting that some open source tools are being built to let anyone use machine learning. For example PredictionIO seems to work on this and they provide a PHP client (the app itself is in java). PredictionIO is an open source machine learning server for software developers to create predictive features, such as personalization, recommendation and content discovery.
  • There are also commercial services offering APIs and machine learning algorithms that may want to integrate with Piwik platform to read the data and provide value to you.

Feel free to post yours thoughts and ideas about what useful things could be done with machine learning applied to Piwik data...

Contributor guide

Use machine learning to create better insights · matomo-org/matomo#5959 | Good First Issue