You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/cognitive-services/QnAMaker/Concepts/best-practices.md
+9-2Lines changed: 9 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -8,7 +8,7 @@ manager: nitinme
8
8
ms.service: cognitive-services
9
9
ms.subservice: qna-maker
10
10
ms.topic: conceptual
11
-
ms.date: 01/10/2020
11
+
ms.date: 01/27/2020
12
12
ms.author: diberry
13
13
ms.custom: seodec18
14
14
---
@@ -25,7 +25,7 @@ In general, FAQ pages should be stand-alone and not combined with other informat
25
25
26
26
### Configuring multi-turn
27
27
28
-
Create your knowledge base with multi-turn extraction enabled. If your knowledge base does or should support question hierarchy, this hierarchy can be extracted from the document or created after the document is extracted.
28
+
[Create your knowledge base](../how-to/multiturn-conversation.md#create-a-multi-turn-conversation-from-a-documents-structure) with multi-turn extraction enabled. If your knowledge base does or should support question hierarchy, this hierarchy can be extracted from the document or created after the document is extracted.
29
29
30
30
<!--is this a global setting that can only be configured at kb creation time? -->
31
31
@@ -46,9 +46,16 @@ Your user may enter questions with either a conversational style of text, `How d
46
46
47
47
The best answers are simple answers but not too simple. Do not use answers such as `yes` and `no`. If your answer should link to other sources or provide a rich experience with media and links, use [metadata tagging](../how-to/edit-knowledge-base.md#add-metadata) to distinguish between answers, then [submit the query](../how-to/metadata-generateanswer-usage.md#generateanswer-request-configuration) with metadata tags in the `strictFilters` property to get the correct answer version.
48
48
49
+
|Answer|Follup-up prompts|
50
+
|--|--|
51
+
|Power down the Surface laptop with the power button on the keyboard.|* Key-combinations to sleep, shut down, and restart.<br>* How to hard-boot a Surface laptop<br>* How to change the BIOS for a Surface laptop<br>* Differences between sleep, shut down and restart|
52
+
|Customer service is available via phone, Skype, and text message 24 hours a day.|* Contact information for sales.<br> * Office and store locations and hours for an in-person visit.<br> * Accessories for a Surface laptop.|
53
+
49
54
## Chit-Chat
50
55
Add chit-chat to your bot, to make your bot more conversational and engaging, with low effort. You can easily add chit-chat data sets from pre-defined personalities when creating your KB, and change them at any time. Learn how to [add chit-chat to your KB](../How-To/chit-chat-knowledge-base.md).
51
56
57
+
Chit-chat is supported in [many languages](../how-to/chit-chat-knowledge-base?branch=pr-en-us-100699#language-support).
58
+
52
59
### Choosing a personality
53
60
Chit-chat is supported for several predefined personalities:
Copy file name to clipboardExpand all lines: articles/cognitive-services/QnAMaker/Concepts/query-knowledge-base.md
+11-9Lines changed: 11 additions & 9 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -25,7 +25,7 @@ The process is explained in the following table.
25
25
|2|QnA Maker preprocesses the user query with language detection, spellers, and word breakers.|
26
26
|3|This preprocessing is taken to alter the user query for the best search results.|
27
27
|4|This altered query is sent to an Azure Cognitive Search Index, which receives the `top` number of results. If the correct answer isn't in these results, increase the value of `top` slightly. Generally, a value of 10 for `top` works in 90% of queries.|
28
-
|5QnA Maker uses syntactic and semantic based featurization to determine the similarity between the user query and the fetched QnA results.|
28
+
|5|QnA Maker uses syntactic and semantic based featurization to determine the similarity between the user query and the fetched QnA results.|
29
29
|6|The machine-learned ranker model uses the different features, from step 5, to determine the confidence scores and the new ranking order.|
30
30
|7|The new results are returned to the client application in ranked order.|
31
31
|||
@@ -47,8 +47,8 @@ A user query is the question that the end user asks of the knowledge base, such
47
47
"scoreThreshold": 20,
48
48
"strictFilters": [
49
49
{
50
-
"name": "feature",
51
-
"value": "collaboration"
50
+
"name": "QuestionType",
51
+
"value": "Support"
52
52
}],
53
53
"userId": "sd53lsY="
54
54
}
@@ -67,20 +67,22 @@ The HTTP response is the answer retrieved from the knowledge base, based on the
67
67
"answers": [
68
68
{
69
69
"questions": [
70
-
"What is the closing time?"
70
+
"How do I add a collaborator to my app?",
71
+
"What access control is provided for the app?",
72
+
"How do I find user management and security?"
71
73
],
72
-
"answer": "10.30 PM",
74
+
"answer": "Use the Azure portal to add a collaborator using Access Control (IAM)",
0 commit comments