The language models are build with the latest version of redirects, disambiguations, and instance-types artifacts, downloaded from the DBpedia Databus. The Catalan, Finish, Lithuanian, and Romanian languages was integrated to the creation model list.

This tool now uses the wikistatsextractor by the great folks over at DiffBot. This means: no more Hadoop and Pig! Running the biggest model (English) takes around 2h on a single machine with around 32GB of RAM. We recommend running this script on an SSD with around 100GB of free space.

• Git
• Maven 3

You can use this tool for creating models of DBpedia Spotlight in your language.

1. docker run -it dbpediaspotlight/model-quickstarter bash

   Generate the models outside the container

   - If you want to generate the models outside the container, just map volumes for the folders `/model-quickstarter/wdir`, `/model-quickstarter/data` and `/model-quickstarter/models`, e.g.

    docker run -v /home/user/data/model/wdir:/model-quickstarter/wdir -v /home/user/data/model/data:/model-quickstarter/data -v /home/user/data/model/models:/model-quickstarter/models -it dbpediaspotlight/model-quickstarter bash
   

2. cd model-quickstarter/

3. Copy & paste one of the following commands to begin the corresponding language model creation process.

You can find pre-built datasets created using the model-quickstarter here:

• Pre-built Spotlight models
• Raw counts

If you use the current (statistical version) of DBpedia Spotlight or the data/models created using this repository, please cite the following paper.

@inproceedings{isem2013daiber,
  title = {Improving Efficiency and Accuracy in Multilingual Entity Extraction},
  author = {Joachim Daiber and Max Jakob and Chris Hokamp and Pablo N. Mendes},
  year = {2013},
  booktitle = {Proceedings of the 9th International Conference on Semantic Systems (I-Semantics)}
}