[Cogsvcs] LUIS - log=true, active learning

Author: diberry

articles/cognitive-services/LUIS/get-started-portal-deploy-app.md

@@ -8,7 +8,7 @@ manager: nitinme
 ms.service: cognitive-services
 ms.subservice: language-understanding
 ms.topic: quickstart
-ms.date: 12/17/2019
+ms.date: 01/27/2020
 ms.author: diberry
 #Customer intent: As a new user, I want to deploy a LUIS app in the LUIS portal so I can understand the process of putting the model on the prediction endpoint.
 ---
@@ -41,7 +41,7 @@ You create the prediction endpoint resource in the Azure portal. This resource s
    |Authoring location|**West US**|The Azure region for authoring.|
    |Authoring pricing tier|**F0**|The default pricing tier for authoring.|
    |Runtime location|**West US**|The Azure region for prediction endpoint queries.|
-   |Runtime pricing tier|**S0**|This pricing tier provides for a high-traffic websites.|
+   |Runtime pricing tier|**S0**|This pricing tier provides for high-traffic websites.|
    | | | |
 
 
@@ -71,6 +71,9 @@ Every time you create a new resource for LUIS, you need to assign the resource t
 
 1. Find the new row in the table for the new prediction resource and copy the endpoint URL. It's correctly constructed to make an `HTTP GET` request to the LUIS API endpoint for a prediction.
 
+> [!TIP]
+> If you intend to use Active learning to improve your LUIS app, select **Change query parameters** and select **Save logs**. This action changes the example URL by adding the `log=true` querystring parameter. Copy and use the changed example query URL when making prediction queries to the runtime endpoint.
+
 ## Train the app
 
 [!INCLUDE [LUIS How to Train steps](includes/howto-train.md)]

articles/cognitive-services/LUIS/luis-how-to-review-endpoint-utterances.md

@@ -1,15 +1,15 @@
 ---
 title: Review user utterances - LUIS
 titleSuffix: Azure Cognitive Services
-description: Review utterances captured by active learning to select intent and mark entities for read-world utterances; accept changes, train and publish.
+description: Review utterances captured by active learning to select intent and mark entities for read-world utterances; accept changes, train, and publish.
 services: cognitive-services
 author: diberry
 manager: nitinme
 ms.custom: seodec18
 ms.service: cognitive-services
 ms.subservice: language-understanding
 ms.topic: conceptual
-ms.date: 01/23/2020
+ms.date: 01/27/2020
 ms.author: diberry
 ---
 
@@ -23,6 +23,18 @@ The process of reviewing endpoint utterances for correct predictions is called [
 
 To enable active learning, you must log user queries. This is accomplished by calling the [endpoint query](luis-get-started-create-app.md#query-the-v3-api-prediction-endpoint) with the `log=true` querystring parameter and value.
 
+Use the LUIS portal to construct the correct endpoint query.
+
+1. In the preview LUIS portal, select your app from the list of apps.
+1. Go to the **Manage** section, then select **Azure resources**.
+1. For the assigned prediction resource, select **Change query parameters**.
+1. Toggle **Save logs** then save by selecting **Done**.
+
+    > [!div class="mx-imgBorder"]
+    > ![Use LUIS portal to save logs, which is required for active learning.](./media/luis-tutorial-review-endpoint-utterances/luis-portal-manage-azure-resource-save-logs.png)
+
+     This action changes the example URL by adding the `log=true` querystring parameter. Copy and use the changed example query URL when making prediction queries to the runtime endpoint.
+
 ## Correct intent predictions to align utterances
 
 Each utterance has a suggested intent displayed in the **Aligned intent** column.