[
https://issues.apache.org/jira/browse/STANBOL-1040?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Rupert Westenthaler updated STANBOL-1040:
-----------------------------------------
Description:
EntityScores are used to represent the importance of an entity within a
knowledge base.
This provider will rank
Administrative regions
==========
based on the Feature Code
* PC*: 1.0
* ADM1*: 0.5
* ADM2*: 0.25
* ADM3*: 0.125
* ADM4*: 0.062
* ADMD*: 0.062
* ADM5*: 0.031
* others: 0.062
Populated Places
======
based on the population
* population: the population of the current Entity
* maxPopulation: 10.000.000
* minPopulation: 1.000
* base = 10.000
* normPopulation: max(minPopulain,min(maxPopulation,population))/base
* ranking: ln(1+normPopulation)/ln(1+(maxPopulation/base))
Some Examples:
10m: 1.0
5m: 0.9
1m: 0.67
500k: 0.57
100k: 0.35
50k: 0.26
10k: 0.1
5k: 0.059
1k: 0.014
Other Feature Classes
==========
Those are currently not supported and will get an default value of 0.1 what is
similar to an ADM3 and cities with a population of ~10.000
was:
EntityScores are used to represent the importance of an entity within a
knowledge base.
This provider will rank
* Administrative regions: 1.0
* Populated places based on the population
* places > 1million : 1.0
* places < 1million in the range [0.2, 1.0] based on the natural log of the
population
* places without population : 0.2
* all other feature classes: 0.3
> Simple Geonames EntityScoreProvider
> -----------------------------------
>
> Key: STANBOL-1040
> URL: https://issues.apache.org/jira/browse/STANBOL-1040
> Project: Stanbol
> Issue Type: New Feature
> Components: Entityhub
> Reporter: Rupert Westenthaler
> Assignee: Rupert Westenthaler
> Priority: Minor
>
> EntityScores are used to represent the importance of an entity within a
> knowledge base.
> This provider will rank
> Administrative regions
> ==========
> based on the Feature Code
> * PC*: 1.0
> * ADM1*: 0.5
> * ADM2*: 0.25
> * ADM3*: 0.125
> * ADM4*: 0.062
> * ADMD*: 0.062
> * ADM5*: 0.031
> * others: 0.062
> Populated Places
> ======
> based on the population
> * population: the population of the current Entity
> * maxPopulation: 10.000.000
> * minPopulation: 1.000
> * base = 10.000
> * normPopulation: max(minPopulain,min(maxPopulation,population))/base
> * ranking: ln(1+normPopulation)/ln(1+(maxPopulation/base))
> Some Examples:
> 10m: 1.0
> 5m: 0.9
> 1m: 0.67
> 500k: 0.57
> 100k: 0.35
> 50k: 0.26
> 10k: 0.1
> 5k: 0.059
> 1k: 0.014
> Other Feature Classes
> ==========
> Those are currently not supported and will get an default value of 0.1 what
> is similar to an ADM3 and cities with a population of ~10.000
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
