> For the complete documentation index, see [llms.txt](https://docs.allganize.ai/allganize-alli-works-userguide-en/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.allganize.ai/allganize-alli-works-userguide-en/q-and-a/q-and-a-settings.md).
# Q\&A settings
In the Q\&A settings menu, you can adjust Alli's Q\&A search results by changing various threshold values. For example, you can make it more flexible to show 10 question/answer candidates for a customer's question, or have it compare not only the question but also the answer content to find search results. Let's learn how to do that.
You can find the QA settings menu by going to Knowledge Base QA. Click the icon located just below the QA Training button.
Now let's go over each setting. Before we start, you need to understand the concept of 'confidence'. If you search QA using the search box in the screenshot above, you can see that each search result displays a 'base confidence' and an 'adjusted confidence' as shown below.
Base confidence is a metric that shows how similar Alli's AI judges the search query and the QA to be. Adjusted confidence is a confidence value that is adjusted according to the customer's selection frequency.
**Question similarity threshold**
QA search only shows customers the search results that have an adjusted confidence higher than the set question similarity threshold.
The first screenshot shows the result when the question similarity threshold is set to 0.55 (55%), and the second screenshot shows the result when it is set to 1.0. As you can see, the preview results changed according to the adjusted confidence values.
**Direct answer similarity threshold setting**
If a specific QA's adjusted confidence is high enough, Alli will show the answer directly rather than presenting question/answer candidates as button options. The direct answer similarity threshold is the reference value for this: Alli will only provide a direct answer when there is a question/answer with an adjusted confidence higher than the value set here.
This setting is another way to control whether to give a direct answer. If this option is on, Alli will not give a direct answer when the customer's question does not end with a question mark.
**Answer similarity threshold setting**
Alli can determine search results by comparing not only the question part but also the answer part of registered QAs with the customer's search query. This setting determines how much the similarity of the answer should be considered.
**Number of question candidates setting**
A simple setting: for example, if this is set to 5, up to 5 question/answer candidates will be shown. However, even in this case, only candidates with adjusted confidence higher than the question similarity threshold are shown. For example, if the setting is 5 but only 3 candidates meet or exceed the question similarity threshold, only 3 candidates will be shown in the search results.
**Keyword matching priority setting**
This value determines how much keywords affect search results. When set higher, search results that contain matching keywords will be given higher priority. Because it's difficult to predict exactly how results will change based only on this setting, as mentioned above, we recommend adjusting the value while checking how the preview results change to decide on a value.
**Confidence adjustment speed setting**
This value controls how much a customer's selection affects the adjusted confidence. When set higher, the adjusted confidence changes more drastically based on customer selections. For more details, **refer to the guide**.
Pressing the reset button here will initialize all adjusted confidences and return them to the base confidence values.
**Other settings**
The first option determines whether to compare the search query against the answer content as well. This option must be enabled for the answer similarity threshold setting above to work.
If the second option is enabled, part of the answer will be shown when presenting QA candidates to the customer.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.allganize.ai/allganize-alli-works-userguide-en/q-and-a/q-and-a-settings.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.