You are here:
Filtering Using Containers and Attributes
On the segmentation canvas, you can define rules to filter your population to a targeted group. You can create and combine filters in different ways for different results. When you define a filter using a related attribute, a container is added to the canvas with applicable rule options. The related attribute’s data model object (DMO) becomes the container object.
If you add other related attributes to the container, they act on the same data row as the container object. The attribute library displays objects up to five relationships away from the container. You can include the related attributes of any of these objects in the container.
However, attributes in separate containers have no relationship to each other and filter your segment separately.
For each container on the canvas, you can calculate the segment population attributed to that container (container population). This helps you see how much each container narrows your segment. The count is shown in a ribbon below the container and is calculated only when you trigger it, and not shown automatically when you add the container.
Scenario 2: If you place the attribute “yellow” in one container, and the attribute “scarf” in another container, the query engine looks for customers who purchased any yellow product and purchased a scarf of any color.
If your attributes are unrelated in your data model, you can’t combine them in a container. Review your data model to understand which attributes are ingested.
- Aggregation
The first step in defining your filter criteria is to decide how you want to quantify the results, such as an average or sum, also known as aggregation. For example, you use different criteria to find five sales order products as compared to five sales orders. Use the container object to aggregate your data. The Data Model Object (DMO) of a related attribute in a container becomes the container object. - Group, Rank, and Limit Segment Audience
The grouping, ranking, and limiting rules in Data 360 Segmentation adds advanced, multi-level filtering to the segment builder. After your standard filters are applied, you can use sequential logic to group, rank, and limit your segment population, creating more structured and balanced audiences. - Hierarchical Aggregation in Segment Filters
Hierarchical aggregation provides a comprehensive view of your data by rolling up revenues for accounts and their subsidiaries. You can then define more precise segment filters based on an aggregate value across multiple levels. - Container Path
Some containers have multiple paths back to the segmentation entity. When a container has multiple access paths, you can select a path to help Data 360 understand how to build your segment. - Issues with Using Cyclic Paths in Segment Attributes
When you choose a DMO more than one time while selecting a related attribute, you create a cyclic path. Cyclic paths are relationships that use cyclic behavior (a→b→a). Cyclic paths degrade segment performance and can lead to count or publish failures. To avoid these issues, Data 360 prevents you from creating cyclic paths through Segment Builder by default. - Event Time in Data 360
An event date is a time-based attribute in a DMO that is defined at the data stream level during ingestion. A DMO can have multiple event dates. The Event Time Field applies only to DMOs designated with the Engagement category. - Create a Container for Segmentation
When a filter is defined using a related attribute, a container is created on the canvas with options to define the rules that you need. You can include up to 20 filters within a container. - Use Vector Filters in Data 360 Segments (Beta)
Vector filters reduce reliance on exact keyword matches and manual data organization, allowing you to create accurate segments efficiently. This feature uses NLP algorithms to sort through unstructured data in segment creation. It supports filtering both structured and unstructured data within a single interface and a segment. - Currency Data Type in Segmentation
You can use the currency data type to convert currency when performing aggregations and creating the required groups in segmentation filters.
