A package implementing #AutoCal was published on #PyPI:

🔗 https://pypi.org/project/matchain/

The source sits on #GitHub in a repository administered by the main author of the paper (who is not in the #Fediverse so far):

🔗 https://github.com/ae3000/matchain

Also, there's no implementation in #CommonLisp so far. 😇

4/4

🌺

🏷️ #InstanceMatching #RecordLinkage #OntologyMatching #ArtificialIntelligence #MatChain #WorldAvatar #DigitalTwin #WebSem #LinkedData #KnowledgeGraph #MachineLearning #DeepLearning #Python #Lisp

From the abstract:

›We also select an unsupervised state-of-the-art matcher from the field of #DeepLearning for a thorough comparison.

Our results show that neither #AutoCal nor the state-of-the-art matcher is superior regarding matching quality while AutoCal has only moderate hardware requirements and runs 2.7 to 60 times faster.‹

3/4

🌺

🏷️ #InstanceMatching #RecordLinkage #OntologyMatching #ArtificialIntelligence #MatChain #PyPI #WorldAvatar #DigitalTwin #WebSem #LinkedData #KnowledgeGraph

From the abstract:

›We introduce #AutoCal, a new #InstanceMatcher which does not require #LabelledData and runs out of the box for a wide range of domains without tuning method-specific parameters.

AutoCal achieves results competitive to recently proposed unsupervised matchers from the field of #MachineLearning.‹

2/4

🌺

🏷️ #InstanceMatching #RecordLinkage #OntologyMatching #ArtificialIntelligence #MatChain #PyPI #WorldAvatar #DigitalTwin #Python #WebSem #LinkedData #KnowledgeGraph

May I kindly draw your attention to this scientific paper in the #JournalOfWebSemantics, since my fate was to read many versions of it and to comment extensively:

›A simple and efficient approach to #unsupervised #InstanceMatching and its application to #LinkedData of #PowerPlants‹

→ https://doi.org/10.1016/j.websem.2024.100815

1/4

🌺

🏷️ #MachineLearning #InstanceMatching #RecordLinkage #OntologyMatching #ArtificialIntelligence #AutoCal #MatChain #PyPI #WorldAvatar #DigitalTwin #Python #WebSem #KnowledgeGraph