# Contents Why extract solutions from the Ticket Comments?

Why extract solutions from the Ticket Comments?

Setup Guide for Knowledge Generation

     Step-by-Step Guide to Setup the Knowledge Generation

     Loading the JSON file to show knowledge fields and values on the Knowledge Mapping page 

     Knowledge Generation configuration setup

     Kick off Knowledge Generation

Generated Knowledge Data source

Knowledge Generation & Similarity Check Logic

     Knowledge Generation Data processing

    Similarity check logic

Tickets Considered for Subsequent Runs

Analyze KB Article Generation Results

     AI Generated Documents from Bulk Tickets

     AI Generated Document from Single Ticket

AI Generated Document Details-Side Panel

Functionality of Generated Documents that are Recrawled

Similar Documents Page

    Functional Details

<br>

Agents frequently disregard resolution notes for closed tickets.&#x20;

.To address this concern, we have devised a solution that extracts vital information from diverse ticket comments. This capability empowers us to create precise and elaborate solutions that form a sturdy knowledge base. This resource can then be utilized by other agents to efficiently resolve similar issues in less time.&#x20;

Our primary objective is to enhance overall knowledge coverage. By granting customers easy access to a wide range of solutions and information, we simplify the process of creating a knowledge base from scratch. This, in turn, improves customer experience.&#x20;

Imagine having a comprehensive knowledge base at customers' fingertips, enriched with insights derived from the collective experiences of agents. With our solution, customers won't have to rely solely on resolution notes. The valuable information extracted from comments offers a holistic view of the ticket resolution process. By implementing our solution, customers can significantly reduce the time spent on information retrieval and knowledge base creation. Instead, they can gain access to a diverse repository of knowledge encompassing various issues and resolutions.&#x20;

<br>

Step-by-Step Guide to Setup the Knowledge Generation&#x20;

Below are the required steps before running the Knowledge Generation:&#x20;

1. APP Creation&#x20;
2. Tickets Ingestion&#x20;
3. Workflow Creation&#x20;
4. Event Setup&#x20;
5. Loading the JSON file to show knowledge fields and values on the Knowledge Mapping page.&#x20;
6. KB Gen Configuration Setup&#x20;
7. Job Run&#x20;

<br>

APP Creation&#x20;

Make sure you have created an Aisera application or bot and added a Data Source to it.&#x20;

Review the Knowledge Base Article Generation Policy for your bot, and choose an AI model for your generation job (you might need your Aisera team to do this for you&#x20;

depending on permissions).&#x20;

Click View Policy under the Knowledge Generation Policy section. A configuration window opens, displaying the prompt settings used for KB Article generation.&#x20;

From the dropdown, select the supported AI model based on your agreement during onboarding or sales discussions.&#x20;

Supported Models Include:&#x20;

GPT-4o&#x20;

LLaMA 3.3&#x20;

LLaMA 3.1&#x20;

Tickets Ingestion&#x20;

When you choose a Data Source for your application from the Aisera Admin UI, the fields to transfer data from the Aisera Gen AI platform to your data source have already been mapped.&#x20;

Review and customize the field mapping for your application, as needed. For KB Generation, while ingesting the tickets below conditions are mandatory.&#x20;

Need a minimum of 40K tickets to form a homogeneous cluster, less tickets more chances 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

5&#x20;

<br>

to we get heterogenous clusters where you see slight different topics merge into one.&#x20;

Please ensure that all relevant comment fields are ingested from the source system to Aisera. KB Generation considers comments to be of utmost importance, so it's crucial to include all types of comments present in the source system. Keep in mind that each customer may maintain different fields to store their comments, so it's essential to understand these fields and ingest them accordingly&#x20;

During the ingestion of comments, it is crucial to map various comments fields to the ‘CaseComment text’ field in the Data source from where we ingested the tickets (Contact connectors team for custom script).&#x20;

We can also map the ticket fields where customers store issue-related information. These fields can be mapped to the comments section to enrich the knowledge details.&#x20;

We support the comments attached to ‘CaseComment text’ and ‘Resolution Notes’.&#x20;

Below are the columns that must be properly mapped and are mandatory for KB Generation.&#x20;

Tickets should be filtered for KB Generation in the Generate Knowledge module, or you can ingest the tickets into DS and use DS without any additional filters&#x20;

Choose the right Type of tickets - Incidents, Service Requests etc.&#x20;

Select the tickets that are marked as closed or resolved. This selection in the conditions (configuration) is based on the assumption that closed and resolved tickets tend to have valuable comments that provide meaningful resolutions. However, it's important to note that our system has the capability to scan and extract meaningful resolution notes from all tickets, not just limited to closed or resolved ones. Note: The quality of the resulting Knowledge Document is determined by the quality of the Tickets you provide.&#x20;

Workflow Creation&#x20;

This workflow is used to upload AI-generated documents to the external knowledge management system. 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

6&#x20;

<br>

Please contact the Aisera team to configure this workflow. The workflow configuration is customized based on the target integration system (for example, Salesforce, ServiceNow, or Confluence).&#x20;

Event Setup&#x20;

This is to trigger the workflow when user wish to upload the AI generated document to the external source system.&#x20;

Navigate to AI Automation -> Event Studio (if you dont see Event Studio, go to settings -> configuration -> Feature Flags ->(check) Enable Event Studio (Beta))&#x20;

<br>

Click on ‘New Event’ -> Give the Event Name, mark Trigger Type as ‘internal’ and select ‘Status’ option as ‘Active’ and click on Next.&#x20;

<br>

Select “Data Type” as Knowledge Base Article&#x20;

<br>

“Event Type” as PublishToSOR&#x20;

“Tiggering condition” as ‘Template’ = \<Template of the document which was configured while generating the document> 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

7&#x20;

<br>

In the Event Handler select the workflow which is configured in the above step and select all input parameters shown below and add corresponding mapping values (same as in the screenshot below) and click on ‘OK’.&#x20;

With this step we successfully created the event&#x20;

Loading the JSON file to show knowledge fields and values on the Knowledge Mapping page&#x20;

This is for the preselection requirements for knowledge field mapping when uploading documents to an external knowledge system.&#x20;

For example, if a user wants the Author field to automatically be set as “John Fernandes” during document upload, this can be configured in advance. Once configured, the author’s name will be automatically populated when the document is published to the external 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

8&#x20;

<br>

knowledge system.&#x20;

If such preselection requirements exist, please provide the following details for each field you would like to configure prior to generating the Knowledge Base (KB) articles. These configurations are uploaded into Aisera using a JSON file. The JSON file must be created with the required field details and then uploaded through the Admin UI (refer to the sections below).&#x20;

To create the JSON file, the following information is required:&#x20;

Field Name&#x20;

Field Path&#x20;

Data Type&#x20;

Allowed Values within the Field&#x20;

Selection Type (Single-select or Multi-select)&#x20;

Additionally, please review the technical specifications below regarding the required JSON format structure.&#x20;

<br>

KB Generation Field Mappings&#x20;

External System Fields JSON&#x20;

{&#x20;

"fields": \[&#x20;

{&#x20;

"fieldName": "string", //Field Label which user sees in the UI when mapping for a job.&#x20;

"fieldPath": "string", //&#x20;

"dataType": "string",&#x20;

"externalFieldType": "string"&#x20;

"allowMultiSelect": true,&#x20;

"values": \[&#x20;

"test",&#x20;

"test2"&#x20;

]&#x20;

}&#x20;

]&#x20;

}&#x20;

fieldName: (Required field)&#x20;

This is the label which user sees in the external system and also which user sees in the Aisera UI when mapping for a job.&#x20;

For Salesforce you can find this in Fields & Relationships view under FIELD LABEL column.&#x20;

fieldPath: (Required field)&#x20;

This is the path of the field which is used while uploading this field to external system.&#x20;

For Salesforce you can find this in Fields & Relationships view under FIELD Name column.&#x20;

dataType: (Optional field, Default value is string)&#x20;

This the dataType of the field value.&#x20;

Supported values for this is Integer, Boolean, String.&#x20;

For Integer, Long and Double values use Number as DataType 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

9&#x20;

<br>

externalFieldType: (Optional field)&#x20;

Right now every field is dropdown where user can select one or multiple values based on the allowMultiSelect field, we want to provide support for use cases where user can enter values dynamically for text, date ..e.t.c so to provide custom extensibility we have added this field.&#x20;

allowMultiSelect: (Optional field, Default value is false.)&#x20;

This field is used to specify whether external system supports multiple values for that field or not.&#x20;

Supported values for that field is true or false.&#x20;

values:&#x20;

The list of value supported by external system.&#x20;

Values should of same dataType provided in the above field.&#x20;

*Below are the steps to upload the Jason in Admin UI*&#x20;

Navigation:&#x20;

Go to Settings -> Configuration -> Knowledge Generation.&#x20;

You will see the following screen.&#x20;

External System Type:&#x20;

<br>

This section will appear blank if no files have been uploaded previously.&#x20;

It lists the integrated systems for which JSON files have been uploaded in the past.&#x20;

Add a JSON File:&#x20;

<br>

Click ‘Add a JSON File’ to upload a new file.&#x20;

A popup will appear. Select the integration type and upload the JSON file as per the instructions in the prerequisite section.&#x20;

Adding New Fields or Values:&#x20;

<br>

Any new values added to the existing fields will be added on top of old values&#x20;

Any new field can be added 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

10&#x20;

<br>

In the popup, select the option ‘Add Delta Values from the JSON’ and click ‘OK.’&#x20;

Override Existing Data:&#x20;

<br>

For any existing values to be updated/deleted use this override option.&#x20;

This action will entirely replace the fields and values, and the updated data will be reflected in the Knowledge Generation -> Action -> Configuration -> Knowledge Field Mapping page.&#x20;

Note: User has no capability to remove an existing field from any of the above options provided&#x20;

Supported Data Types:&#x20;

Number Field Types: These fields accept only numerical values. Depending on the 'allowMultiSelect' setting, users can select either single or multiple number values.&#x20;

Text Field Types: These fields accept alpha, numeric, special characters. If multiple values are provided, they appear in a dropdown menu. With 'allowMultiSelect' enabled, users can select multiple options from this dropdown.&#x20;

Date/Time Field Types: These fields provide a date/time picker in the generated knowledge configuration screen, allowing users to select specific dates and times. The values given will be ignored. User should choose the data picker to select the date/time.&#x20;

Picklist: These fields accept alpha, numeric, special characters. Presents a dropdown menu with predefined options, permitting single selection.&#x20;

Picklist (Multi-Select): These fields accept alpha, numeric, special characters. Allows users to select multiple options from a dropdown menu in the configuration screen.&#x20;

Record Type:These fields accept alpha, numeric, special characters. Users can choose from available values in a dropdown menu; these are not displayed as radio buttons in the user interface.&#x20;

Checkbox:These fields accept alpha, numeric, special characters. Displayed as a picklist, enabling users to select either 'true' or 'false'.&#x20;

Email: These fields accept alpha, numeric, special characters. When multiple email addresses are provided, they appear in a dropdown menu. If 'allowMultiSelect' is enabled, users can choose multiple email addresses from this dropdown.&#x20;

Phone: Similarly, if multiple phone numbers are provided, they are listed in a dropdown menu. With 'allowMultiSelect' enabled, users can select multiple phone numbers from this list.These fields accept alpha, numeric, special characters.&#x20;

Limitations:&#x20;

There is no validation for phone numbers and email addresses in their formats. JSON will accept whatever values are provided in the JSON, so please validate before you upload.&#x20;

Important point to note: When you start a job, the system saves the current knowledge field mappings. If these mappings or their values are changed later, such changes won't affect jobs that have already been run. For example, suppose a previous job included a field called "Product Line" with the value "Finance," and this value was used to prefill the "Product Line" field during document publishing. If someone later changes "Finance" to "Financial Module," 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

11&#x20;

<br>

the past job will still reference "Finance." When attempting to publish a document from that past job, the system won't find "Finance" in the updated mappings and will ignore this value. Only the fields and values that still exist in the current configuration will be prefilled during the document upload.&#x20;

With this, you will be able to load the knowledge field names and their corresponding values into Aisera, where they will be pulled and displayed on the Knowledge Generation Configuration -> Knowledge Field Mapping screen, as shown below.&#x20;

Knowledge Generation configuration setup&#x20;

Step 1 - Configuring “Conditions” Parameters&#x20;

The first step for Knowledge Generation is to fill in the Condition Parameters&#x20;

Navigate to Content Generation > AI Generated Documents&#x20;

Select Actions > Configuration.&#x20;

Under Conditions, there is an option called Job Name where you can add the name of the job before running the Knowledge Generation.&#x20;

<br>

You must select a Data Source under Conditions before creating the KB Generation job.&#x20;

2/9/26, 12:29 PM Aisera Product Documentation&#x20;

12&#x20;

<br>

<br>

Choose the fields that you want to use as Filter Conditions. This includes the name of the Ticket Field and the Condition that you want to use to filter the job results.&#x20;

2/9/26, 12:29 PM Aisera Product Documentation&#x20;

13&#x20;

<br>

<br>

Examples of fields you can use include: Data Source, Status, Ticket Type and&#x20;

<br>

Custom Ticket fields.&#x20;

IMPORTANT NOTE: Multiple selections for a single Field Name act as “OR” operations, while multiple Field Conditions for a single Field Name act as “AND” operations.&#x20;

The Data Source selection is mandatory. You can choose multiple Data Sources from which the Tickets should be considered. The rest of the options are optional.&#x20;

If you leave the Max Tickets Count blank, it indicates that all Tickets in the selected Data Source are being considered. If you specify a limit, the tickets will be randomly selected from the chosen Data Sources until the maximum number is reached.&#x20;

2/9/26, 12:29 PM Aisera Product Documentation&#x20;

14&#x20;

<br>

Step 2 - Configuring “Knowledge Field Mapping”&#x20;

This is not mandatory. If there is a requirement to preselect the knowledge fields while uploading the document, select the fields and their corresponding values that were loaded via the JSON file mentioned above.&#x20;

This will allow the mapping of the selected knowledge fields during the document upload to the source system along with the knowledge document.&#x20;

*If the user chooses ‘Auto Publish’, the fields will be automatically considered during the upload.*&#x20;

*If the user selects Manual and Offline, and wishes to upload manually by clicking ‘Publish to SOR’, there is no need to select the mappings again, as they were already configured during knowledge generation*&#x20;

.&#x20;

Assign Values to Fields 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

15&#x20;

<br>

Step 3 - Configuring “Pre- Generation configuration”&#x20;

The second step for Knowledge Generation is setting the Pre- Generation configuration for the resulting knowledge document that you want to create.&#x20;

Navigate to Content Generation >AI Generated Documents.&#x20;

Go to Actions > Configuration under Conditions, add the Job Name, select the Ticket Conditions, and click on Pre-Generation Configuration.&#x20;

Choose one of the Publish options, described below. Each of the options is described below.&#x20;

<br>

Auto Publish: This option automatically uploads documents without the need for manual review to either the Aisera Knowledge Repository or an external knowledge management system such as Salesforce, or ServiceNow.&#x20;

After selecting Auto Publish, two additional sub-options will be displayed. You can choose either one or both.&#x20;

Publish to Aisera: This will automatically upload the document to the Aisera Knowledge Repository for fulfillment purposes.&#x20;

Publish to SOR: This will automatically upload the document to the configured external repository in a draft state, allowing you to review the document in your knowledge repository, before proceeding through their review cycle, and then publishing it. The Aisera GenAI Platform can then consume and use the document as an option for Fulfillment.&#x20;

2/9/26, 12:29 PM Aisera Product Documentation&#x20;

16&#x20;

<br>

Manual Publish: This option generates the document, but does not upload it to the Aisera GenAI Platform or an external knowledge management system. Use this option if a Admin must review the document and can then manually publish it to the Aisera platform or an external knowledge management system.&#x20;

Generated Knowledge Template: Users have the option to select a template for generating knowledge documents. The template can be chosen from the Generated Knowledge Template drop-down menu.&#x20;

Available template options:&#x20;

* KCS&#x20;
* Default&#x20;
* Master KCS&#x20;
* Master Default&#x20;

<br>

If no option is selected, Master Default will be applied by default.&#x20;

Template Behavior&#x20;

* KCS and Default These templates consider one representative ticket within the cluster for knowledge extraction.&#x20;
* Master KCS and Master Default These templates consider up to five representative tickets within the cluster to generate the knowledge document.&#x20;

<br>

Final Step&#x20;

Once all configurations are completed, click OK.&#x20;

When the user clicks OK:&#x20;

* An automatic data source named Generated Knowledge (BOT ID - <#BOT ID>) will be created.&#x20;
* This data source will be automatically attached to the respective bot.&#x20;
* The generated knowledge documents will be stored in this data source.&#x20;

<br>

Failure messages when any of the configurations are missing:&#x20;

Workflow and Events are used to publish documents to the source system. If any of these configurations are missing or incorrect, a clear message will be displayed in the Status column on the Content Generation → AI Generated Documents page.&#x20;

Case 1: If Workflow is not set up correctly:&#x20;

Workflow configuration error: Please review and update the configuration before retrying.&#x20;

Case 2: If Event is not configured correctly&#x20;

The below use case occurs when the user has not created any events to publish the KBs, or the existing events are in an inactive state.&#x20;

* No active event found for the KCS template.&#x20;
* No active event found for the Default template.&#x20;

<br>

The below use case occurs when the event exists and is active, but the triggering 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

17&#x20;

<br>

condition for the template has not been set up.&#x20;

* The template trigger condition is not configured for the active KCS template event.&#x20;
* The template trigger condition is not configured for the active Default template event.&#x20;

<br>

The below use case occurs when more than one active event is configured with the same template condition for example, having two active workflows using the "KCS" template. This is not allowed. At any given time, only one active event should exist per template, such as "Template KCS" or "Template Default."&#x20;

* Multiple active KCS template events found. Only one active event per template is allowed for publishing.&#x20;
* Multiple active Default template events found. Only one active event per template is allowed for publishing.&#x20;

<br>

Case 3: External/API Issues&#x20;

“API failure, please republish.”&#x20;

Kick off ‘Knowledge Generation&#x20;

Once the configurations are set, click on the Generate Knowledge button beside the&#x20;

Action button.&#x20;

You will get an acknowledgement saying the job has triggered and is in progress.&#x20;

You can monitor the status by clicking on View Job in the acknowledgement or by navigating to Settings-> System Jobs -> Generic Jobs -> Search for KB Generation.&#x20;

The Generate Knowledge button is access-controlled. Users must have the “Generate Knowledge” write privilege to view and run the job by clicking the Generate Knowledge button 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

18&#x20;

<br>

Generated Knowledge Data source&#x20;

When the job runs, a data source named “Generated Knowledge \<BotID>” is automatically created to store all AI-generated documents. This data source is specific to the corresponding bot.&#x20;

The Generated Knowledge Data Source cannot be detached from the bot, meaning that you don't have the ability to remove the Data Source from the application.&#x20;

This Data Source cannot be re-used for other applications. You can not attach this Data Source to other applications.&#x20;

<br>

Knowledge Generation & Similarity Check Logic&#x20;

Let’s understand what happens when user click on ‘Generate knowledge’ button.&#x20;

With the click of the ‘Generate Knowledge’ button, the below two process takes place -&#x20;

Generate knowledge for the cluster representatives, followed by&#x20;

Verifying the similarity of the representatives compared with the committed knowledge (in other words, comparing knowledge repository that are available for downstream consumption)&#x20;

knowledge generation Data processing 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

19&#x20;

<br>

Step 1: Ticket in the data source selected in the configuration&#x20;

Tickets are considered from the Data source selected in the Action → configuration in the Generated knowledge page.&#x20;

Step 2: Preprocessing&#x20;

The following checks are performed during preprocessing:&#x20;

Title Validation: The ticket title must not be null or an empty string.&#x20;

Content Check: Either the comment or resolution notes must contain text and must not be null or empty.&#x20;

Timestamp Validation: Comments without timestamps are eliminated.&#x20;

Filtering: Tickets marked as "Good" or "Very Good" from previous clustering runs are filtered out.&#x20;

Check if any of the prompt and model has been changed, if yes, then all bad quality tickets though it is not updated will be reconsidered. If no, then only the bad quality tickets that are updated will be considered.&#x20;

<br>

Step 3: Resolution Classification&#x20;

This step extracts key attributes:&#x20;

Ticket Quality Classification:&#x20;

Very Good: The ticket contains a solution and receives user feedback or instructions.&#x20;

Good Quality : The Ticket contains solutions but not having users feedback.&#x20;

Bad: The ticket lacks a solution or contains irrelevant information.&#x20;

Uunclassified: This is a case where GPT was unable to process the ticket due to model capacity limitations, excessive ticket size, or failure to generate a response in the correct format, among other internal GPT issues.&#x20;

<br>

Resolution Category: Identifies the category under which the ticket's resolution falls. The fixed categories include:&#x20;

Cannot Be Self-Resolved – Internal Fix Needed: Issues arising from internal company systems that require specific personnel intervention like Developer.&#x20;

No Resolution Found / Agent Support Required to Resolve: Issues that the user can partially resolve but necessitate assistance from an agent for certain steps,&#x20;

2/9/26, 12:29 PM Aisera Product Documentation&#x20;

20&#x20;

<br>

<br>

such as granting permissions etc also this category will have the tickets that has no resolution steps.&#x20;

◦ Issue Reviewed in Other Channels & No Resolution Captured: Cases where the issue is discussed offline via teams, slack etc and no solution has been captured.&#x20;

◦ Resolution Without Feedback: Situations where a solution is provided, but has not confirmation that the issue is resolved.&#x20;

◦ Resolution with Feedback: A situation where a solution is provided, and has confirmation that the issue is resolved.&#x20;

<br>

Quality Classification Reason:&#x20;

Provides the rationale for assigning a particular quality category to the ticket. example say for 'Resolution with Feedback" - The agent provided a resolution action which the user followed and confirmed that it resolved the issue.&#x20;

Resolution Identification: Identifies specific comments containing resolution steps (e.g., comment 5, 9 etc).&#x20;

Issue Summarization: Extracts the intent of the issue by summarizing the ticket title and description.&#x20;

Resolution Summarization: Extracts the intent of the resolution from the resolution steps.&#x20;

<br>

Step 4: Ticket Clustering&#x20;

This step organizes tickets into meaningful clusters:&#x20;

Master Cluster Formation: Groups tickets based on the summarized issue.&#x20;

Sub-cluster Formation: Further refined clusters based on resolution summarization. (Summarization of issue + Resolution)&#x20;

Step 5: Representative Extraction&#x20;

It will pick the representative ticket among the sub cluster by considering the below points.&#x20;

Medoid Selection&#x20;

Good/Very Good&#x20;

Resolution Verbosity&#x20;

Step 6: Cluster Evaluation&#x20;

The purpose of this step is to reevaluate the cluster representative to ensure it closely represents the entire cluster&#x20;

Step 8: Inlyer Analyzer&#x20;

To improve the coverage and quality of resolution steps, we have introduced Multi-Resolution Extraction. This enhancement involves analyzing multiple representative tickets within a cluster to generate a richer and more comprehensive knowledge document. This step further refines each cluster by analyzing summarized issues and resolutions, splitting them into smaller, tightly aligned further groups, and selecting multiple representatives to generate more accurate and targeted knowledge documents.&#x20;

How Does It Help?&#x20;

Enables visibility of multiple resolutions related to the same issue/topic.&#x20;

Consolidates various resolution approaches into a single document, helping reviewers view different resolution steps at once.&#x20;

Allows reviewers to edit, select, and retain only the most accurate steps.&#x20;

Note: Achieving 100% tight clusters is challenging due to variations in ticket content. 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

21&#x20;

<br>

To improve clustering:&#x20;

We introduced Master and Sub-clustering to tighten the groupings&#x20;

With the goal of incorporating more representative tickets, it's essential that these representatives address similar issues.&#x20;

Hence, sub-clusters are further grouped to select more accurate representatives for document generation.&#x20;

Each granular group requires a minimum of 2 tickets to ensure meaningful document creation. If any group contains only 1 ticket, those will be merged into a single group.&#x20;

Among the resulting smaller groups, the dominant group will be selected and displayed in the UI.&#x20;

Maximum number of representatives that will be picked in each group is 4.&#x20;

Step 7: Document Generation&#x20;

The final step generates a structured knowledge document based on the clustered and summarized information.&#x20;

Similarity check logic&#x20;

The logic will analyze all the cluster representative documents and compare them against the committed knowledge repository available in the knowledge module. If no duplicates are found, they will be automatically approved and committed, and will be directly accessible under the knowledge module for fulfillments. However, if any documents are found to be duplicates, they will not be automatically committed or approved. Users can view the duplicate documents and decide whether to commit or delete the generated document. We will delve deeper into the functional requirements in the next section.&#x20;

Tickets Considered for Subsequent Runs&#x20;

The following tickets will be considered in future Knowledge Generation runs:&#x20;

Bad-quality tickets that have been updated after the most recent KB Generation run&#x20;

Good-quality tickets classified as outliers (i.e., not part of any valid cluster)&#x20;

Newly ingested tickets&#x20;

Tickets released when an AI-generated document is deleted&#x20;

<br>

Analyze KB Article Generation Results&#x20;

There are two ways to generate Knowledge Base articles from the tickets that are attached to the specific application/bot.&#x20;

From Bulk Tickets (Via Job):&#x20;

1. Navigate to Content Generation > Knowledge Generation Clustered Tickets, choose Job Configuration from the Settings pull-down menu and then click the Generate Knowledge button. The results for this option are returned on the Content Generation -> AI Generated Documents -> Bulk Tab&#x20;

<br>

From Single Ticket: 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

22&#x20;

<br>

<br>

1. Start from the Agent Assist application/bot and click the Generate Knowledge from Individual Ticket button.&#x20;

<br>

The results for this option are returned on the Content Generation -> AI Generated Documents -> Single tab.&#x20;

1. From SOR -> Tickets, users can navigate to the Tickets page in multiple ways:&#x20;
2. Go to SOR → Tickets, or&#x20;
3. Click on the Ticket Count in Content Generation → AI Generated Documents → Bulk tab, or&#x20;
4. Click the Data Processing Funnel button and select any purple or grey box, which will navigate to the Tickets List page.&#x20;

<br>

From the Tickets List page, click on a specific ticket to open the Ticket Details page.&#x20;

Purpose of Navigation via Content Generation&#x20;

Navigation through Content Generation helps customers review tickets within a cluster.&#x20;

* If a user is not satisfied with the automatically selected representative ticket, they can manually choose a specific ticket from the cluster and generate knowledge from that single ticket.&#x20;
* Similarly, for Bad-quality tickets, users can select the ticket, add Additional Comments, and generate the document without waiting for IT agents to update the ticket in source system and re-ingest it into Aisera.&#x20;

<br>

This allows users to directly add comments to a single ticket and generate the document immediately. 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

23&#x20;

<br>

Knowledge Clusters&#x20;

AI Generated Documents from Bulk Tickets&#x20;

Filter Options:&#x20;

Search: The system will allow users to search for 'Knowledge Clusters' or 'Knowledge Documents' (both have the similar names). You can input the desired knowledge cluster or document name to retrieve relevant results. If no search result found, the application displays the message, No results found under the Knowledge Document section.&#x20;

Job Run Drop down Filter option: This option should be enabled by default with the 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

24&#x20;

<br>

latest job run. It has the list of last job runs that are executed. Users will have an option to choose the job run and the corresponding clusters should get displayed in the ‘Knowledge Generation’ tab. Format should be in Date followed with time. By default, this option is selected with the latest run date/time details.&#x20;

Filter options: Users can click on the "+" icon and add the following fields.&#x20;

Similar Documents: This is a drop-down menu displaying options for Yes and No: Yes filters the documents in the return list by similarity.&#x20;

No filters the documents in the return by dissimilarity.&#x20;

Similar Documents Count&#x20;

Similar Documents = Yes (count) The count will represent the number of customer documents that this AI-generated document matches with.&#x20;

Example&#x20;

Assume customers have 3 documents:&#x20;

Customer Document 1 → AI-generated docs matched: (1, 2)&#x20;

Customer Document 2 → AI-generated docs matched: (2, 3)&#x20;

Customer Document 3 → AI-generated docs matched: (1, 3)&#x20;

For AI-generated Document 1:&#x20;

It matches with Customer Document 1 and Customer Document 3&#x20;

So, Similar Documents = Yes (2)&#x20;

Behavior on Click&#x20;

When the user clicks Yes (2), they will be taken to the new Similar Documents page.&#x20;

This page will display the crawled documents that match this AI-generated document in this example:&#x20;

Customer Document 1&#x20;

Customer Document 3&#x20;

The Similar Documents count should also be clickable in the side panel of the AI-Generated Document Details page.&#x20;

KB Accuracy Feedback: This allows users to filter and view generated documents based on the feedback they’ve provided. It has below values 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

25&#x20;

<br>

ALL&#x20;

Highly Accurate&#x20;

Moderately Accurate&#x20;

Inaccurate&#x20;

None&#x20;

Published to Repository: This will display the list of External source knowledge repository where the AI generated documents are uploaded example ‘ServieNow’, ‘Salesforce’ etc along with ‘Aisera’ option&#x20;

Status : “Status" column helps filter documents to track their publishing progress to the source system.&#x20;

In progress – Document publishing is in progress.&#x20;

Published – Document has been successfully published.&#x20;

Failed – Publishing was unsuccessful due to an issue. (see section above: Failure messages when any of the configurations are missing&#x20;

<br>

Generated By: This will display the list of users who triggered the Knowledge Generation job for that application.&#x20;

Template: This is a drop down menu displaying options&#x20;

Default&#x20;

KCS&#x20;

Master Default&#x20;

Master KCS&#x20;

<br>

Default: This template is simple with Title, Description, solution. Extracted with one representative ticket in the cluster&#x20;

KCS is a standard template using the Knowledge-Centered Service methodology, which is a process that involves collecting, structuring, reusing, and improving knowledge content to enhance the customer experience. It has Title, issue, cause, Environment, solution. Extracted with one Representative ticket in the cluster.&#x20;

Master KCS sample: Multiple Resolution, Environment, and other sections will be displayed, each containing multiple representative tickets within the cluster not just one. The same behavior applies to the Master Default view.&#x20;

Tags: This section displays the list of tags associated with the document. Tags are shared across Aisera modules, meaning tags created in Tickets, Requests, or Knowledge will be available and visible in this list.&#x20;

Users can add tags to AI-generated knowledge articles to help organize and filter documents more effectively.&#x20;

For example, if the article relates to VPN issues and John is the SME responsible for review, you can add tags such as “VPN Issues” or “John.” These tags can then be used as filters to quickly retrieve relevant documents for further review\.2/9/26, 12:29 PM Aisera Product Documentation&#x20;

26&#x20;

<br>

Knowledge Clusters&#x20;

The bar graph will display the clusters in descending order. Each cluster represents a distinct set of similar Ticket pools.&#x20;

When you click on a cluster, the Recommended Knowledge Document associated with that cluster is filtered in the table column, to keep it consistent with other pages. Users will have an option to choose from the various data visualization options as shown below. There is also a download option.&#x20;

Fields in Tabular Section&#x20;

Knowledge Cluster: Which holds the knowledge cluster name&#x20;

Knowledge Cluster Representatives: This column holds the recommended knowledge document that is generated for the specific cluster. You can click on the Knowledge Document, which will route you to the corresponding Knowledge Document Details page, where you can view the generated document.&#x20;

Tickets: This column displays the count of tickets per cluster, representing groups of tickets related to similar issues.&#x20;

Similar Documents: This column contains the similar documents found for the specific cluster representatives.&#x20;

Yes: Indicates that the cluster representative has similar documents among the committed (alias live) documents available for downstream consumption. These documents will be auto-approved and auto-committed automatically when no duplicates are found.&#x20;

No: Indicates that the cluster representative has no similar documents. These documents will also be auto-approved and auto-committed automatically when no duplicates are found. They will be available for downstream consumption.&#x20;

<br>

For documents with duplicates, users need to review them and can either Publish to SOR, Publish to Aisera as they wish along with delete option if they feel the 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

27&#x20;

<br>

generated document is duplicate or already exist in their system, as shown in the screenshot below.&#x20;

Template: It will display the Template of the document.&#x20;

Published to Repository: It will display where the document has been uploaded, for example ‘ServiceNow’, ‘Aisera’ etc.&#x20;

Status: 'Status' column in the 'Generated Knowledge' tabular section to track document publishing progress to the source system.&#x20;

KB Gen Accuracy Feedback: It holds the feedback provided by the users to see the corresponding feedback for each document. To view this field, simply go to 'Add/Remove Columns' (represented by three dots) and add the field.&#x20;

Users have the option to edit and resubmit their feedback. They can also delete the feedback, which will reset the section, making it appear fresh with all radio options unselected. The user can then choose to add new feedback or ignore it.&#x20;

Date : It holds the Date & Time of the cluster when it has been extracted.&#x20;

Generated By: This will show the user id who triggered the job.&#x20;

Tags: This will show the assigned tags to the specific document.&#x20;

Job Run On: This Shows the job name provided by user in configuration along with the date and time when the document is generated.&#x20;

Action Button&#x20;

In the Actions dropdown, along with Job Configuration, the following options are available:&#x20;

Set Recurring Schedule&#x20;

Publish to Aisera&#x20;

Publish to SOR&#x20;

Delete Document&#x20;

<br>

Publishing AI-Generated Documents&#x20;

Users can publish AI-generated documents either to Aisera or to their external System of Record by selecting single or multiple documents from the AI Generated Documents page. 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

28&#x20;

<br>

<br>

Publish to Aisera:&#x20;

<br>

When a document is published to Aisera&#x20;

The document is stored under SOR → Knowledge within the Aisera module.&#x20;

The document becomes available for indexing.&#x20;

It can be consumed by downstream systems, including bots.&#x20;

This option is particularly useful for customers with a limited knowledge base who want to quickly review and enable content within Aisera for bot consumption, helping reduce ticket inflow.&#x20;

Publish to SOR:&#x20;

<br>

Publishing to SOR is intended for customers who:&#x20;

Have an established review and approval workflow in their external knowledge management system.&#x20;

Prefer to review and approve AI-generated documents externally before making them available.&#x20;

Once the document is approved and published in the external system, Aisera can ingest the content for bot consumption.&#x20;

How to Publish Documents&#x20;

Users can publish documents by:&#x20;

Selecting one or multiple documents on the AI Generated Documents page.&#x20;

Clicking Actions → Publish to SOR or Actions → Publish to Aisera.&#x20;

Publishing can also be performed directly from the Document Details page.&#x20;

Note: Once a document has been published, it cannot be republished.&#x20;

Deleting AI-Generated Documents&#x20;

AI-generated documents can be deleted using single or multi-selection from the AI Generated Documents page.&#x20;

Select one or more documents.&#x20;

Navigate to Actions → Delete.&#x20;

When an AI-generated document is deleted:&#x20;

The associated representative tickets are released.&#x20;

These tickets become eligible for consideration in subsequent runs.&#x20;

Users can also delete a document directly from the Document Details page by clicking the Delete icon.&#x20;

<br>

Automatic Knowledge Base Generation Trigger&#x20;

The system now supports automated recurring Knowledge Base (KB) generation based on user-defined schedules and configurations. Once configured, this process runs automatically without manual intervention.&#x20;

Available Recurrence Options&#x20;

Monthly&#x20;

Bi-Monthly&#x20;

Quarterly&#x20;

<br>

Steps to Configure Recurring KB Generation&#x20;

Navigate to Content Generation → AI Generated Documents. 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

29&#x20;

<br>

Click Actions → Set Recurring Schedule to open the recurring configuration window.&#x20;

Configure the following options.&#x20;

Configuration Details&#x20;

Frequency Selection&#x20;

<br>

Users can select the recurrence frequency:&#x20;

Monthly&#x20;

Bi-Monthly&#x20;

Quarterly&#x20;

Start Date&#x20;

<br>

After selecting the frequency:&#x20;

Users must specify a Start Date.&#x20;

The Start Date determines when the recurring KB generation job begins.&#x20;

Past dates are disabled. Only current or future dates can be selected.&#x20;

Ticket Threshold Setting&#x20;

<br>

To ensure high-quality clustering, a minimum of approximately 40,000 tickets is recommended.&#x20;

Lower ticket volumes may result in:&#x20;

Looser clusters&#x20;

Broader topics&#x20;

Less meaningful document generation&#x20;

However, since not all customers may have 40K tickets for a specific configuration, the system allows document generation with lower ticket volumes.&#x20;

To address this, a Ticket Threshold option is introduced.&#x20;

If Ticket Threshold is Enabled&#x20;

On the scheduled job run date, the system checks the total ticket count.&#x20;

If the ticket count is below the threshold, KB generation will be skipped.&#x20;

The skipped job will not appear in the UI.&#x20;

Note: In upcoming releases, notification or acknowledgment messages will be introduced to inform users when recurring runs are skipped due to the threshold not being met.&#x20;

If Ticket Threshold is Disabled&#x20;

The system ignores the ticket count threshold.&#x20;

All available tickets will be processed when the job is triggered. 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

30&#x20;

<br>

<br>

Additional Configuration&#x20;

<br>

All other configurations remain the same as those available for a regular job run, including:&#x20;

Ticket Conditions&#x20;

Knowledge Field Mapping&#x20;

Pre-Generation Configuration&#x20;

Troubleshooting 0 Cluster scenario&#x20;

There is a possibility that there will not be any similar tickets found resulting in zero clusters. This situation is more likely to occur when a very small number of tickets are inserted as input, where no clusters can be formed. In such cases, even if auto-publish is enabled, no Knowledge Base articles will be automatically published. This is because the absence of clusters indicates a lack of similar tickets, making the generated documents less valuable for the customer, due to the infrequency of the tickets.&#x20;

Additionally, consider a scenario where 1000 tickets are present, but no clusters are found; if auto-publish is enabled, we would end up publishing 1000 documents to the source system, leading to an influx of unwanted documents for the customers, causing concerns.&#x20;

To address this issue, we have implemented the following solution:&#x20;

Generate a cluster for each ticket, assuming that if no cluster is formed, each ticket is considered different. This allows us to create multiple clusters, displaying ticket-specific documents under the 'Recommended Knowledge Document' section.&#x20;

Keep 'Auto Publish' turned OFF, which means that even if the configuration states 'YES,' the document will not be automatically published to the source system."&#x20;

For now, if you need the document to be published, please contact our Customer Success Team. We will help you publish the requested document.&#x20;

Data Processing & Ticket Quality Transparency&#x20;

The KB generation process currently offers limited visibility into ticket selection, making it difficult for customers to understand why only certain tickets are used for KB creation while others are not. This often leads to inquiries, especially when a large volume of tickets is ingested but only a subset is utilized. Additionally, knowledge quality is directly impacted by ticket quality, which can raise concerns about accuracy and completeness. Therefore, providing transparency into ticket quality is essential for evaluating knowledge quality effectively. It also enables our customers to identify poor-quality tickets and use them for continuous improvement, guiding their agents to add better comments for improved knowledge generation.&#x20;

During testing and validation, this lack of transparency frequently requires detailed investigations to understand why specific tickets were selected. Customers and stakeholders may need to manually review numerous ticket comments to trace solution sources and selection criteria. These efforts demand significant time from both technical teams analyzing system logs and support teams conducting manual reviews.&#x20;

To address this, we introduced a ‘Data Processing Funnel’ that provides visibility into each step of the Knowledge Generation process. 2/9/26, 12:29 PM Aisera Product Documentation&#x20;

31&#x20;

<br>

Breakdown of Data Processing Stages&#x20;

Before getting into the details, lets understand the limitations&#x20;

The Data Processing Funnel will not be available for past jobs that have already been executed. (The reason is that these are new calculations and newly added schemas, and the corresponding data will not be available for past runs executed before April 2025).&#x20;

What it means is that we have no meta data extracted already for the past runs, hence we have no data to show which are of good and bad quality and few other meta data etc, to capture them into their data processing stages.&#x20;

<table data-header-hidden><thead><tr><th valign="top"></th><th valign="top"></th><th valign="top"></th></tr></thead><tbody><tr><td valign="top">Since there is no metadata for past runs, when users run KB Generation, we will consider all tickets, regardless of whether they were part of past runs. This is important because we need to add metadata for all tickets. As the consideration of tickets for consecutive jobs increases, we will gradually start eliminating very good and good quality tickets that were part of past runs. Therefore, we recommend rerunning KB Generation for all tickets, even if they have already been processed. (Refer to the ticket consideration in consecutive runs for clarity). Data processing stages </td><td valign="top">Its summary </td><td valign="top">Data Navigation </td></tr><tr><td valign="top">Total Tickets </td><td valign="top">Represents the total number of tickets in the selected ticket data source/'s on the day of execution. </td><td valign="top"><p>Currently the Total tickets will be in disable mode and non clickable </p><p>When Users click on 'Total Tickets' the csv file gets downloaded. </p></td></tr></tbody></table>


---

# Agent Instructions: 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:

```
GET https://docs.aisera.com/aisera-platform/content-generation/contents-why-extract-solutions-from-the-ticket-comments.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.