You are here: Decision Strategy Manager > Managing predictive and text analytics > Text analytics > Creating sentiment analysis models

Creating sentiment analysis models

Sentiment analysis detects the sentiment of a text unit, for example, to find out whether a product review was positive or negative. Knowledge about customers' sentiments can be very important because customers often share their opinions, reactions, and attitudes toward products and services in social media and communicate directly through chat channels.

Prerequisites

To build machine learning models, you must access the analytics center. This can be done by launching the pyDecisionAnalytics portal. For more information see, Access Group form - Completing the Definition tab.

  1. In Designer Studio, click Launch > Analytics Center.
  2. In the Analytics Center work area, click Create and then click Sentiment analysis.
  3. In the Create Sentiment Analysis window, perform the following actions:
    1. Enter the name of the sentiment analysis model.

    2. Select the language of the model.

    3. Click Start.

  4. Perform the following actions:
    1. Prepare data.

    2. Upload data for training and testing.

    3. Define the training and testing sample.

    4. Create the model.

    5. Review the model.

    6. Export the model.