Data Store for learning analytics events

If you want to collect data from social networks such as blogs and Twitter, do not heasitate to visit: https://github.com/jlsantoso/trackers

Related documents/publications/deliverables

• Journal of Universal Computer Science (http://www.jucs.org/): Under review
• weSPOT project (D5.3): http://wespot.net/public-deliverables

Installation

• Install Node JS: https://nodejs.org/
• Download our app: all the modules are also included in GitHub - check modules folder.
• mongodb: modify the URL of your mongodb in the following file: routes/db.js
• Execute the datastore: node start DS_server.js

Architecture

The following diagram highlights some components of the weSPOT architecture and illustrates how they realize tracking user data.

Services definition

The main services defined are:

Get events

• app.get(path.join(context, '/events'), getAllEventsJSONP)
  • You can add all the parameters defined in the Schema (see bellow)

Store events

• app.post(path.join(context, '/event'), storeEvent);
  • Push single events
• app.post(path.join(context, '/events'), storeEvents);
  • Push array of events

If you want to avoid the cross-domain problem with JavaScript, you can use the following service and add your host to the list of allowed hosts:

• app.post(path.join(context, '/eventc'), storeEventc);

Schema definition

see file routes/db.js

The schema definition is a simplification of xAPI: https://github.com/adlnet/xAPI-Spec/blob/master/xAPI.md

The schema is defined as:

var eventSchema = new Schema({

  username: String,

  verb: String,

  starttime: Date,

  endtime: Date,

  target: String,

  object: String,

  context: Schema.Types.Mixed,

  location: Schema.Types.Mixed,

  originalrequest: Schema.Types.Mixed

},{collection:"events"});