BigCloneEval
Description
BigCloneEval is a framework for benchmarking clone detection recall using BigCloneBench. It makes it very easy to evaluate your clone detection tool against BigCloneBench. It automates the detection and evaluation steps, including automatically handling a tool's scalability constraints when executed for our large benchmark. The recall evaluation experiment is highly customization, including a plug-in architecture. The experiment code is open-source, and can be modified and re-used for your experiment with accreditation and the sharing of your modifications with the community.
BigCloneBench
BigCloneBench is a collection of over 8 million validated clones in the large Java inter-project repository IJaDataset-2.0. BigCloneBench was built by mining IJaDataset for clones of particular functionalities. It contains both inter-project and intra-project clones of the four primary clone types. Every clone is a true clone by their semantic similarity (a common functionality). The clones span the entire range of syntactical similarity.
The full BigCloneBench can be found here, and can be used as a basis for clone detection and software studies. BigCloneEval uses a special version if BigCloneBench packaged for ease of use when measuring and comparing clone detection recall.
Distribution
BigCloneEval is distributed via a github repository, which makes it easy to pull updates. The github page is available here. The BigCloneBench and IJaDataset distributions (tailored for BigCloneEval) are also needed. See the download section. There is also a VM version available with BigCloneEval setup. It also includes a walkthrough of a demonstration of BigCloneEval with the NiCad clone detector.
Downloads
BigCloneEval: download (2016-06-20)
BigCloneBench: download (2016-06-20)
IJaDatsaet: download (2016-06-20)
Video Demonstration
This video demonstration introduced BigCloneEval, and shows its basic usage for a standard experiment. For advanced usage, and for documentation on the commands and customization of the recall evaluation experiment, please refer to the readme on the github page. For best viewing, watch the video fullscreen in the maximum resolution.
BigCloneEval is a framework for benchmarking clone detection recall using BigCloneBench. It makes it very easy to evaluate your clone detection tool against BigCloneBench. It automates the detection and evaluation steps, including automatically handling a tool's scalability constraints when executed for our large benchmark. The recall evaluation experiment is highly customization, including a plug-in architecture. The experiment code is open-source, and can be modified and re-used for your experiment with accreditation and the sharing of your modifications with the community.
BigCloneBench
BigCloneBench is a collection of over 8 million validated clones in the large Java inter-project repository IJaDataset-2.0. BigCloneBench was built by mining IJaDataset for clones of particular functionalities. It contains both inter-project and intra-project clones of the four primary clone types. Every clone is a true clone by their semantic similarity (a common functionality). The clones span the entire range of syntactical similarity.
The full BigCloneBench can be found here, and can be used as a basis for clone detection and software studies. BigCloneEval uses a special version if BigCloneBench packaged for ease of use when measuring and comparing clone detection recall.
Distribution
BigCloneEval is distributed via a github repository, which makes it easy to pull updates. The github page is available here. The BigCloneBench and IJaDataset distributions (tailored for BigCloneEval) are also needed. See the download section. There is also a VM version available with BigCloneEval setup. It also includes a walkthrough of a demonstration of BigCloneEval with the NiCad clone detector.
Downloads
BigCloneEval: download (2016-06-20)
BigCloneBench: download (2016-06-20)
IJaDatsaet: download (2016-06-20)
Video Demonstration
This video demonstration introduced BigCloneEval, and shows its basic usage for a standard experiment. For advanced usage, and for documentation on the commands and customization of the recall evaluation experiment, please refer to the readme on the github page. For best viewing, watch the video fullscreen in the maximum resolution.
