<style> .chartjs-size-monitor { display: none !important; height: 0 !important; overflow: hidden !important; }p { margin: 0; }span.fr-emoticon.fr-emoticon-img { background-repeat: no-repeat !important; font-size: inherit; height: 1em; width: 1em; min-height: 20px; min-width: 20px; display: inline-block; margin: -0.1em 0.1em 0.1em; line-height: 1; vertical-align: middle; } span.fr-emoticon { font-weight: normal; font-family: "Apple Color Emoji", "Segoe UI Emoji", "NotoColorEmoji", "Segoe UI Symbol", "Android Emoji", "EmojiSymbols"; display: inline; line-height: 0; } blockquote { border-left: solid 2px #5e35b1; color: #5e35b1; margin-left:0; padding-left:5px;}blockquote blockquote{ border-color: #00bcd4; color: #00bcd4;}blockquote blockquote blockquote{ border-color: #43a047; color: #43a047;} table.grid{ border-collapse: collapse;} table.grid td, table.grid th { border: 1px solid #ddd;} .fr-fic.fr-dib{ display: block; margin: 5px auto;}.fr-fic.fr-dib.fr-fir{ text-align: right; margin: 5px 0 5px auto;}.fr-fic.fr-dib.fr-fil{ text-align: left; margin: 5px auto 5px 0;}.fr-fic.fr-dii{ float: none; margin: 5px auto;}.fr-fic.fr-dii.fr-fil{ float: left; margin: 5px auto;}.fr-fic.fr-dii.fr-fir{ float: right; margin: 5px auto;}img.fr-dib.fr-fir { margin-right: 0; text-align: right;}img.fr-dib.fr-fil { margin-left: 0; text-align: left;}img.fr-dib { margin: 5px auto; display: block; float: none;}img.fr-bordered { box-sizing: content-box; border: solid 5px #CCC;}img.fr-shadow { box-shadow: 10px 10px 5px 0px #cccccc;}img.fr-rounded { border-radius: 10px; -moz-border-radius: 10px; -webkit-border-radius: 10px; -moz-background-clip: padding; -webkit-background-clip: padding-box; background-clip: padding-box;}</style><p data-pasted="true">Configuration > AI > General Settings</p><p><br></p>
<style> .chartjs-size-monitor { display: none !important; height: 0 !important; overflow: hidden !important; }p { margin: 0; }span.fr-emoticon.fr-emoticon-img { background-repeat: no-repeat !important; font-size: inherit; height: 1em; width: 1em; min-height: 20px; min-width: 20px; display: inline-block; margin: -0.1em 0.1em 0.1em; line-height: 1; vertical-align: middle; } span.fr-emoticon { font-weight: normal; font-family: "Apple Color Emoji", "Segoe UI Emoji", "NotoColorEmoji", "Segoe UI Symbol", "Android Emoji", "EmojiSymbols"; display: inline; line-height: 0; } blockquote { border-left: solid 2px #5e35b1; color: #5e35b1; margin-left:0; padding-left:5px;}blockquote blockquote{ border-color: #00bcd4; color: #00bcd4;}blockquote blockquote blockquote{ border-color: #43a047; color: #43a047;} table.grid{ border-collapse: collapse;} table.grid td, table.grid th { border: 1px solid #ddd;} .fr-fic.fr-dib{ display: block; margin: 5px auto;}.fr-fic.fr-dib.fr-fir{ text-align: right; margin: 5px 0 5px auto;}.fr-fic.fr-dib.fr-fil{ text-align: left; margin: 5px auto 5px 0;}.fr-fic.fr-dii{ float: none; margin: 5px auto;}.fr-fic.fr-dii.fr-fil{ float: left; margin: 5px auto;}.fr-fic.fr-dii.fr-fir{ float: right; margin: 5px auto;}img.fr-dib.fr-fir { margin-right: 0; text-align: right;}img.fr-dib.fr-fil { margin-left: 0; text-align: left;}img.fr-dib { margin: 5px auto; display: block; float: none;}img.fr-bordered { box-sizing: content-box; border: solid 5px #CCC;}img.fr-shadow { box-shadow: 10px 10px 5px 0px #cccccc;}img.fr-rounded { border-radius: 10px; -moz-border-radius: 10px; -webkit-border-radius: 10px; -moz-background-clip: padding; -webkit-background-clip: padding-box; background-clip: padding-box;}</style><style>
p {
margin: 0; font-family: "Poppins", sans-serif, "Roboto";
}
span.fr-emoticon.fr-emoticon-img {
background-repeat: no-repeat !important; font-size: inherit; height: 1em; width: 1em; min-height: 20px; min-width: 20px; display: inline-block; margin: -0.1em 0.1em 0.1em; line-height: 1; vertical-align: middle;
}
span.fr-emoticon {
font-weight: normal; font-family: "Apple Color Emoji", "Segoe UI Emoji", "NotoColorEmoji", "Segoe UI Symbol", "Android Emoji", "EmojiSymbols"; display: inline; line-height: 0;
}
blockquote {
border-left: solid 2px #5e35b1; color: #5e35b1; margin-left: 0; padding-left: 5px;
}
blockquote blockquote {
border-color: #00bcd4; color: #00bcd4;
}
blockquote blockquote blockquote {
border-color: #43a047; color: #43a047;
}
table.grid {
border-collapse: collapse;
}
table.grid td,
table.grid th {
border: 1px solid #ddd;
}
.fr-fic.fr-dib {
display: block; margin: 5px auto;
}
.fr-fic.fr-dib.fr-fir {
text-align: right; margin: 5px 0 5px auto;
}
.fr-fic.fr-dib.fr-fil {
text-align: left; margin: 5px auto 5px 0;
}
.fr-fic.fr-dii {
float: none; margin: 5px auto;
}
.fr-fic.fr-dii.fr-fil {
float: left; margin: 5px auto;
}
.fr-fic.fr-dii.fr-fir {
float: right; margin: 5px auto;
}
.styled-table thead tr th:first-child {
width: 28em; min-width: 28em; max-width: 28em; word-break: break-word;
}
.styled-table thead tr th:nth-child(2) {
width: 8em; min-width: 8em; max-width: 8em; word-break: break-word;
}
.styled-table thead tr th:nth-child(3) {
width: 8em; min-width: 8em; max-width: 8em; word-break: break-word;
}
.styled-table thead tr th:nth-child(4) {
width: 8em; min-width: 8em; max-width: 8em; word-break: break-word;
}
.styled-table thead tr th:nth-child(5) {
width: 8em; min-width: 8em; max-width: 8em; word-break: break-word;
}
.styled-table {
border-radius: 5px;border-collapse: collapse; margin: 25px 0; font-size: 1rem; font-family: "Poppins", sans-serif, "Roboto"; width: 100%; box-shadow: 0 0 20px rgba(0, 0, 0, 0.15);
}
.styled-table thead tr {
text-align: left;border-radius: 5px;
}
.styled-table th {
border-radius: 5px; padding: 12px 15px;
}
.styled-table td {
padding: 12px 15px;
}
.styled-table tbody tr {
border-bottom: 1px solid #dddddd;
}
.styled-table tbody tr:nth-of-type(even) {
background-color: #fafafa;
}
.styled-table tbody tr:last-of-type {
border-bottom: 2px solid $orcolor;
}
</style><p><strong><span style="font-size: 18pt;">Setup</span></strong></p><table class="styled-table grid" style="width: 100%; margin-left: calc(0%);"><tbody><tr><td style="width: 33.3333%; text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Field</span></strong></td><td style="width: 14.4246%; text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Type</span></strong></td><td style="width: 52.1935%; text-align: center; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Description</span></strong></td></tr><tr><td style="width: 33.3333%;" id="ai_connection_type"><p>Default AI Connection</p></td><td style="width: 14.4246%;">Radio List</td><td style="width: 52.1935%; text-align: left;">This determines which AI connection is used. You will need to enter credentials if using Azure/Own OpenAI. (On-Prem customers will not have the option to use the 'Default Halo Connection'). </td></tr></tbody></table><p><br></p><p><strong><span style="font-size: 18pt;">Ticket Embeddings</span></strong></p><table class="styled-table grid" style="width: 100%; margin-left: calc(0%); height: 350px;"><tbody><tr style="height: 38.8333px;"><td style="width: 33.3333%; text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Field</span></strong></td><td style="width: 14.4279%; text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Type</span></strong></td><td style="width: 52.1935%; text-align: center; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Description</span></strong></td></tr><tr style="height: 57.8333px;"><td style="width: 33.3333%;" id="ticket_embeddings_db">Vector search database<br></td><td style="width: 14.4279%;">Radio List</td><td style="width: 52.1935%; text-align: left;">This determines the database used when vector search takes place for ticket matching. The search process involves comparing the embedding scores of the tickets, but the similarity of the scores will differ based on the database used. If using the Azure AI Search (recommended) option you will need to enter your Azure credentials.<br></td></tr><tr style="height: 57.8333px;"><td style="width: 33.3333%;" id="use_embedding_scores_openai"><p id="isPasted">Enable Ticket embeddings and similarity matching</p></td><td style="width: 14.4279%;">Checkbox</td><td style="width: 52.1935%; text-align: left;">When enabled, an embedding will be created for each new ticket upon creation. The system then searches this embedding against existing ones to identify the most similar tickets, ranked by similarity score. The top matches can be used to power AI Suggestions to automatically populate fields and resolutions based on past ticket data.<br></td></tr><tr style="height: 57.8333px;"><td style="width: 33.3333%;" id="ticket_embeddings_method"><p id="isPasted">Ticket matching and AI insights method</p></td><td style="width: 14.4279%;">Single Select</td><td style="width: 52.1935%; text-align: left;">This determines what ticket matching and AI insights are based on. The integration runbooks option allows for more customisation but it is not advised unless you already have an extensive knowledge of runbooks. </td></tr><tr style="height: 79.8333px;"><td style="width: 33.3333%;" id="vector_score_minimum"><p id="isPasted">Minimum vector match score (Tickets)</p></td><td style="width: 14.4279%;">Integer</td><td style="width: 52.1935%; text-align: left;">This determines what the minimum similarity score must be between two Tickets' embedding scores in order for them to be matched as similar. It is a score between 0 and 1 with the standard being 0.8. The higher the score, the stricter the matches, which will return fewer but more accurate matches. A lower score will cast a wider net, returning more results but also more potential for outliers.</td></tr><tr style="height: 57.8333px;" id="vectorisetickets"><td style="width: 33.3333%;">Index Tickets</td><td style="width: 14.4279%;">Button</td><td style="width: 52.1935%; text-align: left;">Here, you can schedule tickets to be indexed. Indexing tickets will re-generate embedding scores for tickets already in the system. It is recommended this is completed when first enabling embedding scores as this will improve initial matching. It is also recommended to do this after changing the matching method, vector database or embedding field used.<br></td></tr><tr><td style="width: 33.3333%;" id="deleteindex_halo-tickets">Delete Ticket Vectors</td><td style="width: 14.4279%;">Button</td><td style="width: 52.1935%; text-align: left;">When used, all vector scores (embeddings) created for tickets will be cleared from your chosen vector search database. Used when <a href="https://www.usehalo.com/guides/2673" target="_blank" rel="noopener noreferrer">switching embedding models.</a> Only available when connecting your own vector search database </td></tr></tbody></table><p><br></p><p><strong><span style="font-size: 18pt;">Article and Service Embeddings</span></strong></p><table class="styled-table grid" style="width: 100%; margin-left: calc(0%); height: 975px;"><tbody><tr><td style="width: 33.3333%; text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Field</span></strong></td><td style="width: 14.4246%; text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Type</span></strong></td><td style="width: 52.1935%; text-align: center; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Description</span></strong></td></tr><tr><td style="width: 33.3333%;" id="ai_knowledge_search_method"><p>Vector search database</p></td><td style="width: 14.4246%;">Radio List</td><td style="width: 52.1935%; text-align: left;">This determines the database used when vector search takes place for AI knowledge search. The search process involves comparing the embedding scores of the tickets, but the similarity of the scores will differ based on the database used. functionality is only available using 'Azure AI Search' database, for this you will need to enter your Azure credentials. </td></tr><tr><td style="width: 33.3333%;" id="textsplit_chunk_size"><p id="isPasted">Chunk Size</p></td><td style="width: 14.4246%;">Integer</td><td style="width: 52.1935%; text-align: left;">This determines the chunk size that is searched.</td></tr><tr><td style="width: 33.3333%;" id="textsplit_chunk_overlap"><p id="isPasted">Chunk Overlap</p></td><td style="width: 14.4246%;">Integer </td><td style="width: 52.1935%; text-align: left;"><p id="isPasted">This determines the overlap of chunk size that is searched. </p></td></tr><tr><td style="width: 33.3333%;" id="vector_score_knowledge_minimum"><p id="isPasted">Minimum vector match score (Knowledge)</p></td><td style="width: 14.4246%;">Button</td><td style="width: 52.1935%; text-align: left;">This determines what the minimum vector score between an article and the search prompt must be in order for the article to be returned in the search. A higher vector score indicates greater similarity between the search prompt and the article. <br></td></tr><tr style="height: 65px;"><td style="width: 33.3333%;" id="restrict_indexing_by_faqlist"><p id="isPasted">Restrict which Articles can be indexed based on their FAQ List</p></td><td style="width: 14.4246%;">Checkbox<br></td><td style="width: 52.1935%; text-align: left;"><p id="isPasted">When checked, you will be able to restrict which articles can be indexed based on their FAQ list. An additional option will appear against each FAQ list to determine if articles in this FAQ list can be indexed. </p></td></tr><tr><td style="width: 33.3333%;" id="restrict_indexing_by_service_category"><p data-pasted="true">Restrict which Service Catalogue can be indexed based on their Service Category</p></td><td style="width: 14.4246%;">Checkbox<br></td><td style="width: 52.1935%; text-align: left;"><strong>(v2.242+)</strong> When enabled, services can be excluded from indexing based on their service category. Against each service category you can determine whether to allow indexing. For a service to be eligible for indexing, it must be directly under a category with indexing enabled. <br></td></tr><tr><td style="width: 33.3333%;" id="include_request_details_in_service_index"><p data-pasted="true">Include Service Request Details and Incident Details when indexing services</p></td><td style="width: 14.4246%;">Checkbox<br></td><td style="width: 52.1935%; text-align: left;"><strong>(v2.242+) </strong>When enabled, request and incident details on a service will be included as part of the embeddings. Meaning they are accounted for in AI search. If a request or incident label matches the search term, the parent service will appear in the results.<br></td></tr><tr><td style="width: 33.3333%;" id="_index_kbs_now"><p id="isPasted">Index Articles</p></td><td style="width: 14.4246%;">Button</td><td style="width: 52.1935%; text-align: left;"><p id="isPasted">Here, you can schedule articles to be indexed. Indexing articles will re-generate embedding scores for articles already in the system. It is recommended this is completed when first enabling article suggestions/knowledge search. </p></td></tr><tr><td style="width: 33.3333%;" id="deleteindex_halo-kb-articles">Delete Article Vectors</td><td style="width: 14.4246%;">Button</td><td style="width: 52.1935%; text-align: left;">When used, all vector scores (embeddings) created for articles will be cleared from your chosen vector search database. Used when <a href="https://www.usehalo.com/guides/2673" target="_blank" rel="noopener noreferrer">switching embedding models</a>. Only available when connecting your own vector search database.<br></td></tr><tr style="height: 85px;"><td style="width: 33.3333%;" id="_index_services_now"><p id="isPasted">Index Service Catalogue</p></td><td style="width: 14.4246%;">Button</td><td style="width: 52.1935%; text-align: left;"><p>Here, you can schedule services in the service catalogue to be indexed. Indexing services will re-generate embedding scores for services already in the system. It is recommended this is completed when first enabling article suggestions/knowledge search. </p></td></tr><tr><td style="width: 33.3333%;" id="deleteindex_halo-services">Delete Service Vectors</td><td style="width: 14.4246%;">Button</td><td style="width: 52.1935%; text-align: left;">When used, all vector scores (embeddings) created for services will be cleared from your chosen vector search database. Used when <a href="https://www.usehalo.com/guides/2673" target="_blank" rel="noopener noreferrer">switching embedding models. </a>Only available when connecting your own vector search database.<br></td></tr></tbody></table><p><br></p><p><strong id="isPasted"><span style="font-size: 18pt;">AI Search</span></strong></p><p> </p><table class="styled-table grid" style="width: 100%; margin-left: calc(0%); height: 241px;"><colgroup><col style="width: 33.3467%;"></colgroup> <colgroup><col style="width: 14.4289%;"></colgroup> <colgroup><col style="width: 52.2245%;"></colgroup><tbody><tr><td style="text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Field</span></strong></td><td style="text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Type</span></strong></td><td style="text-align: center; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Description</span></strong></td></tr><tr style="height: 47px;"><td id="enable_ai_search_default"><p>Enable AI Search by default on search screens</p></td><td>Checkbox</td><td style="text-align: left;">When enabled, all search screens will use AI search by default. AI search can still be disabled manually against each screen upon search. <br></td></tr><tr><td id="enable_user_ai_kb_and_service_search">Allow Users to use AI Search for Articles and Service Catalogue</td><td>Checkbox</td><td style="text-align: left;"><strong>(v2.238+)</strong> When enabled, Users will be able to use the AI search functionality to find articles and services. <br></td></tr><tr><td id="include_article_tags_in_ai_search">Also search Article tags when using AI search</td><td>Checkbox<br></td><td style="text-align: left;">When enabled, the search will include article tags, and will show articles with matching tags at the top of the results list.<br></td></tr><tr><td id="ai_kb_fuzzy_search">Use AI to correct spelling for Knowledge Base and Service searches (Fuzzy Search)</td><td>Checkbox</td><td style="text-align: left;"><strong>(v2.238+)</strong> When enabled, The AI Search will use fuzzy searching to correct spelling for Knowledge Base and Service searches. </td></tr></tbody></table><p><br></p><p><br></p><p><strong id="isPasted"><span style="font-size: 18pt;">Logs</span></strong></p><table class="styled-table grid" style="width: 100%; margin-left: calc(0%);"><tbody><tr><td style="width: 33.3333%; text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Field</span></strong></td><td style="width: 14.4246%; text-align: left; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Type</span></strong></td><td style="width: 52.1935%; text-align: center; background-color: rgb(0, 204, 248);"><strong><span style="color: rgb(255, 255, 255); font-size: 12pt;">Description</span></strong></td></tr><tr><td style="width: 33.3333%;" id="openAILogs"><p id="isPasted">Open AI</p></td><td style="width: 14.4246%;">Button<br></td><td style="width: 52.1935%; text-align: left;">Shows the request/response logs for calls made to OpenAI.</td></tr><tr><td style="width: 33.3333%;" id="azureOpenAILogs"><p id="isPasted">Azure Open AI</p></td><td style="width: 14.4246%;">Button<br></td><td style="width: 52.1935%; text-align: left;">Shows the request/response logs for calls made to Azure OpenAI.<br></td></tr><tr><td style="width: 33.3333%;" id="azureAISearchLogs"><p id="isPasted">Azure AI Search</p></td><td style="width: 14.4246%;">Button<br></td><td style="width: 52.1935%; text-align: left;">Shows the request/response logs for Azure AI Searches.<br></td></tr></tbody></table>
