This Script is part of the Deprecated Content (Deprecated) Pack.#
Creates a text classifier model using machine learning.
Each line of the text file is sample data, and it has to have a tag.
Script Data#
| Name | Description |
|---|
| Script Type | python |
| Tags | phishing, ml |
| Cortex XSOAR Version | 4.1.0+ |
Inputs#
| Argument Name | Description |
|---|
| trainingFileName | The name of the text file (the format is described in the script details). The file should be attached in the War Room. |
| modelStoreListName | The name of the list to store the model binary (as base64). The list must exist in Cortex XSOAR (with any value). |
| precisionThreshold | The precision threshold. The model will be stored only above this threshold. |
Outputs#
| Path | Description | Type |
|---|
| DBotTextClassifier.TrainDataSize | The number of training samples. | number |
| DBotTextClassifier.ListName | The model list name in Cortex XSOAR. | Unknown |
| DBotTextClassifier.Precision | The model precision (precent). | number |
| DBotTextClassifier.Recall | The model recall (precent). | number |