Search Based Classification

Search Based Classification helps to define your keywords for the customized classes. You can add multiple classes and keywords for defining a customized search-based training model. Based on the given classes and keywords, the model will be trained and can be further used to classify the content from the set of classes and keywords that are defined.

The search based classification works on the text. The pre-trained AI-based model provides the percentage of a partial match for the given set of keywords in the form of a confidence score between 0-100. Here 0 is interpreted as no match and 100 is interpreted as a perfect match. Higher the value between 0-100, the more the chances for the document to fall under the predicted category.

To open Search Based Classification application, navigate to Smart Bots > Classification > Search Based Classification. 

The list of trained models for images is displayed. 

The Search-Based Classifiers displays the following information:

• Classifier Name: Specifies the name of the text classifier model.
• Created Time: Specifies the date and time of the dataset when it was created.
• Description: Specifies the description of the classifier. 
• Last Modified Time: Specifies the latest date and time when the classification model was trained.
• Action: Use the Actions column for performing the following operations: 
  • Delete: Delete the model.
  • Export Classifier: Export the search-based classifier model.
  • Classify Content: Classify the content based on the trained model.
  • Edit:  Edit the classes and keywords of the trained model.