Cortex extends the value of your business’s unique data by enabling anyone to transform that data into Machine Learning Pipelines. The types of data supported by Cortex fall into four categories: events, attributes, items, and ID mappings.
|Events *||Timestamped actions taken by your users.||Customer ABC completes a purchase event at time T.|
|Attributes||Characteristics or traits of your users.||Customer ABC has job title ‘Professor’, and age 49.|
|Items||Metadata attributes for the items that your users interact with, via events.||Item XYZ has category ‘shoes’, and price 49.99.|
|ID Mappings||Associations between one set of IDs and another.||Customer ABC has cookie ID 123.|
In this guide, we’ll provide an overview of Items data and discuss the various ways that it may be ingested into Cortex.
What are items?
Items data provides additional information about the items (e.g. products, content, etc.) that your users interact with, via events. Items data is required in order to generate personalized item recommendations for each user. If you’re not interested in such recommendations, Items data is optional, but encouraged — the more data at Cortex’s disposal, the wider the set of predictions you can make, and the more accurate your predictions will be.
Items data can include any metadata descriptors for the items referenced in your events. For example, if you’re tracking customer purchase behavior via Events data, you may supplement that data with item metadata from a Content Management System (e.g. category, price, brand, etc.).
The below table shows what a sample of Items data might look like for a set of commerce products. Note that the ID used to identify your items must match the one included in your events feed.
When you upload a file like this, Cortex stores the attributes with a timestamp indicating when each was ingested. This allows your pipelines to access a historical record of attribute values over time. For example, by recording that Item 69564645 cost 74.99 January 1st but changed to 79.99 on January 8th, your pipeline will know the exact price at the time of each purchase event.
How do I send Items data into Cortex?
There are two ways that you can ingest Items data into Cortex:
1) Batch File Uploads
Items data can be batch uploaded to Cortex in files exported from, say, your Content Management System. To do this, schedule a recurring file upload into a cloud-based directory hosted by either you or Vidora. Cortex provides flexibility around both the format of these files (CSV, JSON, Parquet, etc.) as well as the type of directory into which they’re uploaded (Amazon S3, Google Cloud, Azure, etc.).
2) Custom Feeds
Vidora also has the ability to pull Items data from a custom feed, such as API, XML, or Pub/Sub. For more information, contact your account manager or firstname.lastname@example.org.
Still have questions? Reach out to email@example.com for more info!