Zalak Parikh:

I am new to Machine Learning so please pardon me for any silly question!

I am using AWS ML multiclass classification but have stuck at one point.

I have following dataset as a training data:

Instance 1:

Section - Question - Answer - overallScore - reward

1 - 1 - Yes - 5.2 - 3/4

1 - 2 - No - 5.2 - 3/4

Instance 2:

Section - Question - Answer - overallScore - reward

1 - 1 - No - 3 - 2/6/8

1 - 2 - No - 3 - 2/6/8

Explanation:

As you can see above, I have a simple section wise questionnaire application. Currently I have data available which is generated manually (reward is given manually based on each answer and overall score as well as section). But now I want to predict reward (which can be multiple that's why I had seperated it with forward slash). Reward is dependent on answer for each question, overall score and section. And except score, all other data is categorical. So my question is how can I implement this scenerio with AWS ML? How can I group Multiple Rows to get single prediction?