Options

How to create a dataset using Excel for predicting price hikes?

Surya69Surya69 Member Posts: 1 Newbie
edited January 2020 in Help
I am participating in the SIH 2020. My problem is to create a predictive analysis for hikes in onion price. I have no idea how to do it. Some help would be nice.
Tagged:

Answers

  • Options
    User111113User111113 Member Posts: 24 Maven
    There are a lot of predictive models available in Rapid Miner to choose from, you should also check if anything fits your criteria from default templates.

    Please provide some more context here in order to resolve your query, may be how your data looks like or what have you tried so far or anything that you think would be relevant for us to answer you.

    Thanks
  • Options
    MarcoBarradasMarcoBarradas Administrator, Employee, RapidMiner Certified Analyst, Member Posts: 272 Unicorn
    @Surya69 depending on what you want to predict there are some options. If you want to predict the value for a certain day you may go for a regression if you would want to predict the value for a window larger than a day you may want to go for a Time Series forecast. That's assuming that you want to predict a value. 
    If you want to predict only if the onion price is going to increase as a binary result you may go to other type of model.

    You said your data is on excel? or you want to create something that interacts with excel to create the prediction.
    If your data is in excel format import it with Read Excel operator I guess you'll have a set of dates and values of onions for a wide range of days. 
    After that you need to create a Time Series analysis problem. Go to RapidMiner Academy and watch the Time Series Course which could help you get started.

    If you need anything else or have some answers for the question I'll be glad to try to help you and good luck with Smart India Hackathon. Maybe https://www.aptusdatalabs.com/ could be interested on helping you they are India Based and a great partner of RapidMiner
Sign In or Register to comment.