Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 8 Next »

Introduction:

This documentation describes about the generation of the batch wise assessment report within the course. for more detail refer this PRD.


Solution -1: Assessment Samza job and Data product


Diagram:


As per the above diagram, Router job should route all ASSESS events into Assessment Samza Job which compute and update the RDBMS table.The assessment data product which read from the table and will generates the assessment report per batch and uploads to azure cloud storage


Disadvantages:

  1. When pipeline is having huge lag,Then report will be wrong   

Solution - 2: API and Data product



As per the above diagram, End user will sync the assess events through api, which will update the database with computed values. The assessment data product which read data from the database and will generate the reports per batch and uploads to azure cloud storage.


API:

API Request
METHOD: POST 
URI: action/data/v3/assess/telemetry
BODY: {
  "id": "sunbird.telemetry",
  "ver": "3.0",
  "ets": 1566884714550,
  "events": [] // Only ASSESS Events
}



Disadvantages:

  1. Need to validate and de-dup the the events.

Table Schema:

History table schema
CREATE TABLE assessment_histroy(
    batch_id: text // Batch identifier.
    user_id: text, // User Identifier.
    attempt_id: index, // Question Attempt identifier, To track the number of attempts on that question.
    course_id: text, // Course Identifier.
    content_id: text, // Content Identifier
    question: jsonb // array[objects] // Question details spec object
    datetime: timestamp // Time stamp, To know last updated time.
    PRIMARY KEY (batch_id,  user_id, content_id)
)


Question object :

Question object structure
question:{
  "qid": "do_957399094343",
  "max_score": 5,
  "score": 2,
  "type" :"mcq/mmcq/ftb/mtf",
  "title": "Choose valid odd number"
  "pass": "Yes"/"No",
  "desc":"Description of the question",
  "duration": 5
}

Challenges:

 

  1. How to capture the attempts? i.e. Number of times the particular user is attempted particular question. 
  2. How to capture the batch-id and course-id



Conclusion:







  • No labels