The first thing that needs to be defined is the entity which will be recommended. In the case of the MovieLens-100k dataset, we know that are entity is called "movie".
Please note that by adding multiple item entities, you are defining multiple domains on an item-level.
By creating a new entity, we can define what kind of meta-data that particular entity has available. In the Uptrendz platform, the following attribute types are supported when the data model for item entities:
| Field Type | Field Subtype | Description |
|---|---|---|
| Categorical Text | Single Value | String value which usually depict some kind of category. Most commonly used for post-filtering recommendation results. |
| Categorical Text | Multiple Values | Array of string values which usually depict some kind of category. Most commonly used for post-filtering recommendation results. |
| Free Text | English | Text that should be processed and utilized for content-based recommendation. Language of the text shoud be English. |
| Free Text | German | Text that should be processed and utilized for content-based recommendation. Language of the text shoud be German. |
| Numeric | Integer | Integer number which usually depict some kind of value (e.g., age of a user). Most commonly used for post-filtering recommendation results. |
| Numeric | Real | Real number which usually depict some kind of value (e.g., price of a product). Most commonly used for post-filtering recommendation results. |
| Date | - | Date information for the respective entity (e.g., created date). Expected format: YYYY-MM-DDThh:mm:ssZ |
Given the MovieLens-100k dataset, the following movie configuration should be defined for the Item API:
| Field Name | Field Type | Field Subtype |
|---|---|---|
| id | Categorical Text | Single Value |
| title | Free Text | English |
| release_date | Date | - |
| genres | Categorical Text | Multiple Values |
Once the Item API is prepared, we can upload movie data using the Data Upload notebook.