MiraLak / activityRecognitionV2

Remake of activity recognition project in Scala.

activityRecognitionV2

Remake of activity recognition project in Scala.

There are two main part:

• creation of a prediction model based on a Random Forest algorithm
• Predict the activity in a realtime using the generated prediction model

This project contains an already generated prediction model so you can try it directly.

Prerequisites

• Cassandra 2.2.2
• Java 8
• Scala 2.11.8

Start the application

First of all, get the project https://github.com/MiraLak/accelerometer-rest-to-cassandra and follow the steps to start it. This project will allows you to collect data and exposes it in a Rest API that will be used to recognize the activity. Then, you can start the app by launching the main class fr.duchess.service.PredictionService for the prediction.

Remember!

Make sure that you use the same network and don't hesitate to update the configurations of cassandra of spark if needed.