ACM SIGMOD 2021
Welcome participants of the ACM SIGMOD 2021 programming contest! This page contains some special information for you to get started more easily with our tool.
Feel free to open an issue in case you find a bug or see space for improvement :)
This tool will help you to compare and evaluate your data matching solutions. You can upload experiment results from your data matching solution and then compare it with a goldstandard, compare two experiment runs with each other or calculate binary metrics like precision or recall.
We are working closely with the contest team to allow for a seamless experience.
To use snowman, you'll have to download the latest release (Github Releases) of
snowman for your specific os. After that you are able to start snowman!
As a first step, we'd suggest for you to start with our section Basic Usage or take a look at our intoductory video below for a guide on how to benchmark and evaluate matching solutions with snowman. You can find an explanation of a basic workflow here or in the video below.
Some contest datasets are already bundled with the release. Select one of them, upload your own experiment result and start benchmarking!
In case you'll want to upgrade later on, simply download the newest release and you're ready to go!
We've included an automatic importer for the gold standard format and validated that the below datasets can be imported out of the box. More datasets will be added as soon as they are released by the contest team.
- NotebookToy: prepackaged dataset X1 and gold standard Y1
- Notebook: prepackaged dataset X2 and gold standard Y2
If you want to upload further datasets which have the same format as the SIGMOD-datasets, you will have to change some default settings
in the dataset-uploader-dialog: Set the
ID Column from
instance_id and set the
Escape character to
". You can then select the dataset file and click on
We wish all participants best of luck!