Search Results for

    Show / Hide Table of Contents

    Creating Product Recommendation model

    Overview

    The product recommendation model produces sets of predictive product recommendations. With the help of this predictive model, you can use the customer transaction data like purchases, returns, and donations to provide the basis for educated product recommendations.

    To create the model

    Follow the below steps to create the product recommendation model:

    1. Go to Predict > Built-in.

    The Predictions page appears.


    Alt text


    1. In Predictions page, click Use this model on Product recommendation.

    The Product recommendation form appears.


    Alt text


    1. Click Get Started.

    The Model name page appears.


    Alt text


    To name the model

    Follow the below steps to name the product recommendation model:

    1. Type a unique name in the Name this model for you text area.
    2. Type a name in the Output table name text area.
    3. Click Save & Proceed.

    imageNote

    The Output table name is the name given to the model table. It will appear on the databases page after running the model.

    The Model preferences page appears.


    Alt text


    To add model preferences

    Follow the below steps to define your preferences to be used while suggesting products:

    1. In the Number of products section, select a value in the Total number of products available for customers to buy text area.
    2. Select an option in the Suggest products customers have recently purchased? text area.
    3. In the Look back window section, select the number of Months in the How far back to look at customer’s activity to use as input to their recommendations text area.
    4. Click Save & Proceed.

    The Required data page appears.


    Alt text


    To add required data

    Follow the below steps to add data to the product recommendation model:

    1. In the Required data page, click Add New.

    The Add Customer Data form appears.


    Alt text


    1. Select Purchase history from the drop-down list.
    2. Select Transaction ID from the drop-down list.
    3. Select Transaction date from the drop-down list.
    4. Select Value of transaction from the drop-down list.
    5. Select Unique product ID from the drop-down list.
    6. Specify the information if this transaction was a return from the drop-down list.
    7. Click Save.

    The table gets created.


    Alt text


    To edit the created table

    In the Required data page, you can click the three horizontal ellipses under the Action column to edit the created table.

    • If required, you can perform one of the below actions:
    If you want to Then
    Edit the created table Select Edit.
    Delete the created table Select Delete.
    • Once done, click Save & Proceed in the Required data page.

    The Data update schedule page appears.


    Alt text


    To add the Data update schedule

    Follow the below steps to schedule your data update:

    1. In the Data updates page, perform one of the following actions:
    To Do
    Import your data automatically every week Select Weekly.
    Import your data automatically every month Select Monthly.
    1. Click Save & Proceed.

    The Review and run page appears.


    Alt text


    To review and run your model

    Follow the below steps to save your Product recommendation model or run it:

    1. In the Review your model details page, click the Edit icon to edit the details.
    2. Once you complete your review or editing, perform one of the below actions:
    To Do
    Save and run your product recommendation model Click Save & Run.
    Save your product recommendation model Click Save & Close.
    • Improve this Doc
    In This Article
    Back to top © SkyPoint Cloud Inc. All rights reserved.