SimilarityMeasure: Stringbased
Developer: Björn Endres
Description
This module determines the similarity of two ontology entities by calculating a simple string alikeness on the labels of these entities. The PhaseLibs ontology model provides these labels with the getName() method. The algorithm is based on a 3-gram matcher by Micheal Sintek.
Characteristics
In order to demonstrate the basic characteristics of this SimilarityMeasure, here some examples:
Frame A name | Frame B name | Measure value |
Analyser | AnaLyseR | 1.00 |
Analyser | Analyzer | 0.50 |
Analyser | Sample-Analysis | 0.59 |
Sample-Analysis | Analyser | 0.30 |
Evaluation/Performance
This measure was used for some basic evaluation tests. You can see the results here, second column.
Specification
Intitialisation
The SimilarityMeasure main class is
de.dfki.km.phaselib.impl.similarities.stringbased.StringBasedSimilarity
Initialisation is straight forward:
new StringBasedSimilarity()
Parameters
none
Dependencies
none
License Issues
TODO