Find Insights: Cluster

Cluster is the perfect landing page, allowing you to quickly, simply, and easily understand what's happening in your data at a high-level.

📘
Cluster
Your automated report on the key themes in your unstructured data.

What does Cluster do?

Clustering groups items so that those in the same group/cluster have meaningful similarities (i.e. specific features or properties).
One-to-one assignment. Meaning, that an individual data point (e.g. response, document, comment) is assigned to one cluster. Under clustering, a data point cannot be assigned to two or more clusters (note for one-to-many relationships, see 'Tagging').
Clustering facilitates informed decision-making by giving significant meaning to data through the identification of different patterns. It is a great tool to unravel hidden patterns in the data.

What comes with Cluster?

Group similar items (comments, responses, images) into meaningful buckets together
Understand the top high level themes in your unstructured data without effort
Explore hidden patterns in your data
Uncover unexpected and emerging themes without bias, by viewing the data without the need for an upfront, fixed taxonomy
Drill down to identify the detail behind each cluster, including what's driving each group, instantly

How does it work?

There are variety of clustering algorithm introduced under AI and data-processing. At Relevance AI, we employ vector-base analysis which relies on state of the art vector representation of data (you can read more about vectors at What are vectors). The next step is to focus on the similarity of the data points in the vector space and the defined criteria to specify the clusters.

Explorer Dashboard in Relevance AI

Explorer is the best tool to study your data and analyse its composing clusters. Explorer is an advance tool not only for data presentation but also for imposing metrics and aggregations. It enables you to perform tasks such as

Identify key themes
Understand what's driving similarity
Compare clusters
Label clusters
Merge clusters
Move documents/responses to different clusters
Search within the data (semantic and word matching)
Apply different filters
Add visualisations
Introduce metrics on the whole data or within clusters
Aggregate data by existing categories within the whole dataset or within single clusters
Export clusters