An approach for predicting defect proneness on unlabeled datasets- Average Clustering and Labeling (ACL).
ACL models get good prediction performance and are comparable to typical supervised learning models in terms of F-measure. ACL offers a viable choice for defect prediction on unlabeled dataset.
This implementation is based in the following paper:
@INPROCEEDINGS{7828414,
author={J. Yang and H. Qian},
booktitle={2016 IEEE 18th International Conference on High Performance Computing and Communications; IEEE 14th International Conference on Smart City; IEEE 2nd International Conference on Data Science and Systems (HPCC/SmartCity/DSS)},
title={Defect Prediction on Unlabeled Datasets by Using Unsupervised Clustering},
year={2016},
volume={},
number={},
pages={465-472},
keywords={data handling;pattern clustering;public domain software;software fault tolerance;unsupervised learning;cross-project defect prediction;CPDP;unlabeled dataset;unsupervised clustering;software engineering;unsupervised learning;transfer learning;data distribution;open source project;Measurement;Predictive models;Supervised learning;Prediction algorithms;Labeling;Unsupervised learning;Software;Defect Prediction;Unlabeled Datasets;Unsupervised Cluetring},
doi={10.1109/HPCC-SmartCity-DSS.2016.0073},
ISSN={},
month={Dec},
}