Temporal Relation Classification

Temporal Relation Classification is a subtask in the field of natural language processing that focuses on classifying the temporal relationships between time entities. This task aims to identify and classify thirteen types of temporal relations between conventional events and time expressions. However, due to the ambiguity in annotation, recent corpora have simplified it into a more explicit subset. By introducing labels such as "no relation" or "ambiguous," the scope of Temporal Relation Classification can be further expanded, enhancing the accuracy and practicality of temporal relation recognition.