Changes between Version 3 and Version 4 of generator_PhaseTab
- Timestamp:
- 09/26/06 11:03:43 (18 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
generator_PhaseTab
v3 v4 6 6 7 7 == Description == 8 The PhaseTabalgorithm is a generic algorithm that employs an arbitrary number of [wiki:SimilarityMeasure SimilarityMeasures] to generate an alignment. The algorithm was originally developed by [mailto:Malte.Kiesel(at)dfki.de Malte Kiesel] and [mailto:Ludger.van_Elst(at)dfki.de Ludger van Elst] and has been presented in the paper referenced [wiki:PhaseLibs here]. It was then adapted for the PhaseLibs project.8 The [wiki:generator_PhaseTab PhaseTab] algorithm is a generic algorithm that employs an arbitrary number of [wiki:SimilarityMeasure SimilarityMeasures] to generate an alignment. The algorithm was originally developed by [mailto:Malte.Kiesel(at)dfki.de Malte Kiesel] and [mailto:Ludger.van_Elst(at)dfki.de Ludger van Elst] and has been presented in the paper referenced [wiki:PhaseLibs here]. It was then adapted for the PhaseLibs project. 9 9 10 The basic principle of the algorithm is the iterative confirmation of the best ranked relation proposals. This approach makes use of a feature of some [wiki:SimilarityMeasure SimilarityMeasures], especially the SimilarityFlooding, which increase their precision when given higher confidences on correct relations.10 The basic principle of the algorithm is the iterative confirmation of the best ranked relation proposals. This approach makes use of a feature of some [wiki:SimilarityMeasure SimilarityMeasures], especially the [wiki:similarity_SimFlood similarity flooding], which increase their precision when given higher confidences on correct relations. 11 11 12 In order to archieve a robust and fair ranking, a [wiki:generator_BordaCount BordaCount] alogrithm is used to identify the {{{n}}} best rated realtion of all SimilarityMeasuresinvolved.12 In order to archieve a robust and fair ranking, a [wiki:generator_BordaCount BordaCount] alogrithm is used to identify the {{{n}}} best rated realtion of all [wiki:SimilarityMeasure SimilarityMeasures] involved. 13 13 14 14 An iteration cycle now consists of three phases: … … 34 34 35 35 == Specification == 36 === Intitialisation ===37 36 38 37 === Parameters === 39 none 38 39 || '''Parameter name''' || '''ValueType''' || '''Default''' || '''Description''' || 40 || PARAM_TAXONOMY_ONLY || Boolean || {{{FALSE}}} || Defines whether only the taxonomy (classses) should be aligned. This defaults to aligning the classes along with the properties.|| 41 || PARAM_THRESHOLD || Double || {{{0.0}}} || This standard parameter forces the algorithm to drop all relations with lesser confidence. However, this parameter must be chosen carefully, since the confidences produced by the [wiki:generator_BordaCount BordaCount] algorithm are no explanatory values. It would probably make more sense, to apply a threshold to each SimilarityMeasure, but this requires for a set of threshold parameters and not just a single one.|| 42 || PARAM_MAXIMAL_ITERATIONS || Integer || {{{64}}} || Defines the number of iteration cycles to be performed. Thus (PARAM_MAXIMAL_ITERATIONS * PARAM_CONFIRMATION_STEP_SIZE) relations eventually be confirmed.|| 43 || PARAM_CONFIRMATION_STEP_SIZE || Integer || {{{5}}} || Defines the number of relations to confirm with each iteration.|| 44 || PARAM_FIX_1_0_RELATIONS || Boolean || {{{TRUE}}} || EXPERIMENTAL: This enables the user to lift the 1.0 restriction, i.e. each iteration the relations with confidence 1.0 are set to 0.9999 to enable the system to review them. However, first test produced poor results.|| 40 45 41 46 === Dependencies === 42 none 47 48 The algorithm uses the [wiki:generator_BordaCount BordaCount] alignment generator and the MatchingAlignment alignment implementation. 43 49 44 50 == License Issues == 45 TODO 51 52 This module is subject to the license the PhaseLibs project is published under.
