Unsupervised learning
A learning method that looks for patterns in data without predefined target answers for the particular training task.
What patterns does unsupervised learning seek?
Unsupervised learning looks for patterns in data without using predefined target answers for the particular training task. One possible task is grouping similar examples; another is condensing many features into a smaller representation. Data preparation, features and the method influence which similarities matter. People must define and assess the task and the subsequent usefulness of the results. A discovered pattern does not automatically come with a valid business interpretation.
In a fictional service project, messages without prior topic labels are to be grouped by content similarity. Employees examine the resulting groups to check whether they bring recurring requests together in an understandable way. Language or similar phrasing might shape the grouping more strongly than the underlying issue. The team therefore plans examples of the same request in different words and different requests in similar words. The groups initially serve investigation. Naming them and potentially using them as working categories require separate assessment; neither priority nor ownership is established by the groups alone.
Sources
Related terms
- Machine learningModels learn patterns from data and apply them to new inputs.
- Large language modelA large language model processes language using learned patterns. Fluent text proves neither factual accuracy nor access to current business data.
- LLM evaluationLLM evaluation tests model outputs against defined tasks and criteria. One good example does not establish overall quality.
- Anomaly detectionMethods that flag unusual observations relative to a defined reference or a learned pattern.
- Supervised learningA learning method that uses inputs together with their corresponding target values or classes as training examples.
- Re-rankingRe-ranking reassesses search candidates already retrieved. Relevance to the query and the limits of the initial selection both matter.
Your project with ASES
Describe your project to us. We will get back to you with an initial assessment.