Node.js Interactive 2015

Node.js and its ecosystem enable fast and easy development of applications. However, data analytics components typically require external tools such as Apache Storm or Spark. In this presentation Blaž Fortuna and Jan Rupnik will present how to use or develop scalable data mining components directly from Node.js using open-source modules. This will be presented through two large-scale use-cases: topic and sentiment classification on Twitter and real-time user profiling. Use-cases cover different data modalities (text, social networks, streams), transformations (feature extraction) and algorithms (classification, recommendation). The core enabler for this are open-source modules that allow Node.js developers to perform fast numeric computation and data processing close to the hardware directly from JavaScript.