Dataset from a Smart Factory for Activity Detection using Temporal Case-based Reasoning

This dataset accompanies the paper "Case-Based Activity Detection from Segmented Internet of Things Data" published in the proceedings of the 33rd International Conference on Case-Based Reasoning (ICCBR) 2025 (DOI: 10.1007/978-3-031-96559-3_29). 

The dataset contains:

storage_process.bpmn and production_process.bpmn: the two process models executed using the Camunda 7 platform to control the smart factory

signatures folder: the signatures for the activities executed by the smart factory to be used as cases in individual files; created based on the annotations from the data analyst; also includes the unprocessed IoT data (file: input-log.jsonl)

iot-data folder: the segmented IoT data of activity executions to be classified by using the proposed approach (filenames contain the expected label as ground truth); also includes the process event logs created by the Camunda 7 platform that were used to segment the IoT data

confusionMatrices.csv and mostSimilarCases.jon: results from evaluating the newly developed TCBR-based approach applied to the IoT data