# AI Model Set Up

An AI recommendation model is set up in three steps:

* **Step 1: Model Configuration** 
  * Configure the Model Name.
  * Choose a domain or specific terms within the domain.
  * Proceed to Step 2.
* **Step 2: Object & Source Selection**
  * Object Type (Schema, Table, Table Columns, etc.)
  * Source Type (Connector or Schema)
  * Refining Source Selection. (Selection of a list of connectors or schema)
  * Configuring the model to run on delta data or all data 
  * Configuring the model to run on Associated objects (Objects with term association or Unassociated objects 
  * Notification preference - Notify steward of term, data object, or both 
  * Proceed to Step 3.
* **Step 3: AI Configuration**
  * Set up Smart Score.
  * Set up other configurations.

To create an AI Model in the OvalEdge application, follow these steps:

1. Navigate to Governance Catalog > Data Classifications Recommendations. 
2. Click on Create Model. A pop-up with a three-step process will appear: Model Configuration, Object & Source Selection, and AI Configuration.

   <figure><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXeHvC6qzRh-giC9Ka4A8OFClCXncka2CdzWauP6apKYsE4wT19gJ2sDpiNtml2O_r72FCAGSTobbhBZGzBWbEhaReAYmjMmTk_RUOG97Q-ayMqPazpcWxPfIm5XG3Xb7-HOGXLpkg?key=UB3jazI03Domow6i7y49wQ" alt=""><figcaption></figcaption></figure>

## Configure Model Settings

Enter the following details and proceed to Object and Source Selection.

| Field Name           | Description                                                                                           |
| -------------------- | ----------------------------------------------------------------------------------------------------- |
| AI Model Name        | Provide the model's name for easy recognition.                                                        |
| AI Model Description | Describe the purpose of the model.                                                                    |
| Domain               | Select a domain. Terms within this domain will be used for recommendations.                           |
| Category             | Choose a category within the domain. Terms in this category will be used for recommendations.         |
| Subcategory          | Choose a subcategory within the category. Terms in this subcategory will be used for recommendations. |
| Terms                | Select one or more terms. These terms will be recommended for the selected data objects.              |

## Object & Source Selection

After configuring the model details, click Continue to define the objects and source for recommendations.

<figure><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXdvpf_azGJNQfewBhT3iaaabVTclHb5NdMFuX9DTkDl9_LErqOUOM30a_HYa63TtjTeJ1cu-uZSPT6lAD4jQAw8ektMw4sbfDQBPpCKs6HkSlef6MyPMmDCCy0vV3xrCN97nz8RVA?key=UB3jazI03Domow6i7y49wQ" alt=""><figcaption></figcaption></figure>

Enter the following details and proceed to AI Configurations.

<table><thead><tr><th width="203">Field Name</th><th>Description</th></tr></thead><tbody><tr><td>Object Type</td><td>Select the data object type that requires recommendations.</td></tr><tr><td>Source Type</td><td>Specify if recommendations are required for the selected data object from a connector or schema.</td></tr><tr><td>Refine Source Selection</td><td><p>Based on the "Connector" or "Schema" selection for the Source Type, specify the connectors or schemas to include.</p><p>Select Objects</p><p>Users can choose the Source Type for which they require recommendations.</p><ul><li>If the Object Type is selected as “Table” and the Source Type is “Connector,” the Refine Source Selection displays all established connectors in the system.</li><li>If the Object Type is selected as “Table” and the Source Type is “Schema,” the Refine Source Selection displays all Schemas related to different Connectors established in the system.</li></ul><p>The system supports multiple source types for the model, with a maximum limit of 20 at any given instance. Search filters such as Connector Name, Connector ID, and Connector Type for Connector Source, and Connector Name and Schema Name for Schema, are enabled to simplify the search and locate relevant Source Types. Users can select Sources by clicking on the corresponding rows.</p></td></tr><tr><td>Run</td><td><p>This section provides two options to define the data processing range used by the AI Model for generating recommendations:</p><ul><li>All Data: Processes all available data for the chosen object (from the connector or schema) to generate recommendations.</li><li>Delta Data: Processes recently crawled data and data not previously analyzed by the AI model since the last run to generate recommendations.</li></ul></td></tr><tr><td>Include Objects</td><td><p>This section presents two options for determining how the AI Model processes data to provide recommendations for the selected object.</p><ul><li>All Objects: Processes all data, regardless of term associations, within the chosen object from the connector or schema.</li><li>Unassociated Objects: Processes only unassociated data objects without any term associations within the chosen object from the connector or schema.</li></ul></td></tr><tr><td>Notification Preference</td><td><p>Specify the notification preference when the recommendations are generated and the available options are </p><ul><li>Notify the Steward of the Term</li><li>Notify the Steward of the Data Object </li><li>Notify both the Steward of the Term and the Steward of the Data Object </li><li>Notify None</li></ul></td></tr></tbody></table>

## AI Configuration

After selecting the Object and Source, click Continue to configure AI parameters for recommendations.

<figure><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXc6f-qy3K6Q9gBXKZs3asXC3TQkZYtW5HFHzwr_XJAsPc_sGHU15_73-e78WDvfWlAUSz48PsQVz48hruA3ik6c2n_Gy3oh8hWjulNk0jPo0gVlh0OJzWPD_TFI5HHvbnmLzCzmtg?key=UB3jazI03Domow6i7y49wQ" alt=""><figcaption></figcaption></figure>

Define AI Score and other configurations for object recommendations, and click Save to create the AI Model.

### AI Score Configurations

<table><thead><tr><th width="383">Configuration Name</th><th>Description</th></tr></thead><tbody><tr><td>Smart Score</td><td><p>Set a smart score value required for a data object to be recommended by AI. Only objects with Smart Scores equal to or above this value will be displayed.</p><p>Default value: 10</p></td></tr><tr><td>Threshold score</td><td><p>Set the threshold score to directly associate a term with a data object without manual acceptance/rejection of Term recommendations. Smart Scores equal to or higher than this value will automatically associate recommended terms.</p><p>Default value:0</p><p>Choose a value higher than your typical AI smart score to capture relevant recommendations. This value should be a positive whole number.</p></td></tr><tr><td>Smart Score Boost for Name Matches</td><td><p>Configure the value to add and boost the Smart Score if the column name matches the term name for AI Recommendations.</p><p>Default value: 50.</p></td></tr><tr><td>Smart score boost for Column Repetition</td><td><p>Set the value to increase the Smart Score and boost it when multiple repeated columns are present in the associated data objects.</p><p>Default value: 0.1.</p></td></tr><tr><td>Name, Data, Pattern Weightage for Smart Score</td><td><p>Assign weight ratios to factors (Name, Data, and Pattern) when calculating the 'Smart Score.'</p><p>Enter weights: Specify the percentage for Name, Data, and Pattern when calculating the Smart Score. The sum of these values should be around 100.</p><p>Example: Name (30): Data (30): Pattern (40) - This gives 30% weight to Name, 30% to Data, and 40% to Pattern.</p><p>These weights are not set by default</p></td></tr><tr><td>Rejected Score weightage</td><td><p>Configure the rejected score weightage percentage for AI recommendations. This weightage determines how much the smart score of a data object should be influenced if the user rejects a recommended term.</p><p>Default value: 20.</p></td></tr></tbody></table>

### **Smart Score Calculation**

The AI recommendation algorithms operate in the backend based on Smart Score, which analyzes the Name, Data, and Pattern of an object to determine the relevance of an object-term relation. The algorithm starts by analyzing the characteristics of the data objects, including their name, metadata, and content. It then examines the patterns in the data, including the frequency and co-occurrence of different terms within the data objects.

Once the Smart Score is calculated, the algorithm generates a list of the most relevant data objects for the given term.

**Smart Score**&#x20;

Smart Score displays scores based on four parameters: Name, Data, Patterns, and Data Types.

* **Name Score:** It is calculated by matching the Name of the column with aggregated Names in the AI Model, which generates a fuzzy score. A name in the model can be repeated more than once (Email: 2); in such cases, we add some weight to that score. Finally, the maximum score among all is considered for the Fuzzy Score.
* **Data Score:** It is calculated by matching the Top values with aggregated top values in the AI Model. The Top Values are identified during profiling.
* **Pattern Score:** The pattern score is displayed based on the data patterns of the column, as they exist in the AI Model.
* **Data Types Score:** The recommendation model considers data types when recommending business glossary terms, improving accuracy. The system compares the data type of a data object (e.g., int, char, varchar) with the data types of objects already associated with a glossary term. If the data types match, the smart score for that term increases by +10.

**Example:** Suppose the glossary term "Transaction Date" is already linked to Date data type columns across the system.

* When the model evaluates a new column, "Order Date" of Date type, it identifies a data type match.
* The smart score for recommending "Transaction Date" increases by +10 points, making it a stronger candidate.

However, if the system evaluates "Transaction ID" (an int column), no data type match occurs, so no additional points are added. This reduces the chance of irrelevant suggestions.

### AI Configuration for Objects&#x20;

<table><thead><tr><th width="369">Configuration Name</th><th>Description</th></tr></thead><tbody><tr><td>Maximum Columns for AI Recommendations</td><td><p>Set the maximum number of table columns for AI Recommendations.</p><p>Default: 50,000</p></td></tr><tr><td>Pattern Recognition</td><td>Enable 'True' to allow the AI Model to recognize data object patterns. Disabling 'False' will prevent pattern recognition.</td></tr><tr><td>Exclude Matching Names</td><td>Enter names to exclude from recommendations. These names may be in data objects or have partial matches.</td></tr><tr><td>Except these Names</td><td>Enter names or partial matches that should not be excluded from recommendations.</td></tr><tr><td>Exclude Tagged Data Objects</td><td>Choose "Yes" to exclude tagged data objects from recommendations. Choose "No" to include all data objects in recommendations.</td></tr><tr><td>Delimiter Configuration</td><td><p>In the Run AI Recommendations pop-up, set the delimiter to separate multiple included/excluded database names before executing the AI Recommendation Job.</p><p>Default value: "","" (comma).</p></td></tr></tbody></table>