# Uni-Target Class Prediction Task

A Uni-Target Class Prediction Task is a class prediction task that is a single-target prediction task (which requires a single value associated to a test case).

**AKA:**Single-Label Classification.**Context:**- It can range from being a Heuristic Unilabel Classification Task to being a Data-Driven Unilabel Classification Task (such as a supervised unilabel classification task).
- …

**Counter-Example(s):****See:**Single-Target Numeric Value Prediction Task.

## References

### 2007

- (Tsoumakas & Katakis, 2007) ⇒ Grigorios Tsoumakas, and Ioannis Katakis. (2007). “Multi-Label Classification: An Overview.” In: International Journal of Data Warehousing and Mining, 3(3). doi:10.4018/jdwm.2007070101
- QUOTE: Traditional single-label classification is concerned with learning from a set of examples that are associated with a single label l from a set of disjoint labels [math]L[/math], [math]\mid L \mid \gt 1[/math]. If [math]\mid L \mid = 2[/math], then the learning problem is called a binary classification problem (or filtering in the case of textual and web data), while if [math]\mid L \mid \gt 2[/math], then it is called a multi-class classification problem.
In multi-label classification, the examples are associated with a set of labels [math]Y ⊆ L[/math]. In the past, multi-label classification was mainly motivated by the tasks of text categorization and medical diagnosis.

- QUOTE: Traditional single-label classification is concerned with learning from a set of examples that are associated with a single label l from a set of disjoint labels [math]L[/math], [math]\mid L \mid \gt 1[/math]. If [math]\mid L \mid = 2[/math], then the learning problem is called a binary classification problem (or filtering in the case of textual and web data), while if [math]\mid L \mid \gt 2[/math], then it is called a multi-class classification problem.