BurgerLinker v0.8.0

This release features a largely reworked BurgerLinker with support for custom data models and post-processing rules, with output conformant to the Persons-in-Context data model, and which abandoned the need to provide the input in HDT format.

Main changes:

* Bumped dependencies to latest version and updated code to match.
* Refactored project structure to make source file hierarchy more sensible.
* Cleaned up (parts of) code for readability purposes.
* Improved source variable naming to reflect actual purpose.
* Reduced repetition: the different Within, Between, and Closure classes
  and functions have been generalized to fit all modi.
* Changed unsecure Schema.org namespace to secure variant.
* Fixed bug in age check that excluded the maximum value.
* Fixed bug with incorrect metadata in closure function output of Within
  data.
* Added feature to provide custom data model for input graphs.
* Added feature to provide custom base namespace for reconstructed
  individuals.
* Added feature to provide custom validation rule set for processess.
* Changed intermediate CSV result structure to ensure consistency between
  different modi.
* Change generation of reconstructed person identifiers to be based on
  person IDs, ensuring consistency between runs. URIs starting with 'I-'
  pertain between runs; those starting with 'U-' can change.
* Added header to links-individuals file.
* Added feature to allow appending new data to existing data store.
* Change input parameter names to better convey their purpose.
* Added feature to provide basic query functionality.
* Added feature to execute all functions after one another if none is
  specified.
* Changed progress bar to activity indicator.
* Cleaned up messages to standard output.
* Added native support for text-encoded RDF format (e.g. N-Triples and
  N-Quads). This removes the need to first convert the input data to
  HDT format. Instead, an on-disk triple store is created and queried.
* Added support for (remote) SPARQL endpoint as input.
* Fixed bug in timeline consistency check of parents that checked age
  instead of age difference.
* Changes output format of closure funtion to conform to PiCo
  PersonReconstructed objects.
* Added CSV-to-RDF conversion scripts for CIV and PiCo data models.
* Removed the need for (event and person) identifiers in favour of IRIs.
* Removed duplicated entries from output of closure function that were
  caused by aggregating all attributes of linked individuals.
* Updated and extended README.
* Many minor changes.

burgerLinker v0.7

This version fixes a problem caused by external libraries when running burgerLinker in terminal

burgerLinker v0.6

This version fixes a bug in the closure function and updates the HDT library to the latest version (v 3.0.2)

burgerLinker v0.5

This version supports one-to-one matching, using the --singleInd flag.

Example: for matching newborns to deceased individuals (with a maximum Levenshtein distance of 2) without the requirement of matching their parents as well, you can use the following command:

java -jar burgerLinker.jar --function Within_B_D --inputData dataDirectory/myCivilRegistries.hdt --outputDir . --maxLev 2 --singleInd

burgerLinker v0.4

Closure function covers all types of links

burgerLinker v0.3

Bug fixes, and added the following tool parameters:

--ignoreDate (optional, default = False) add this flag without a value (i.e. True) for ignoring the date consistency check before saving a link. By default, the tool only saves links that are temporally consistent (e.g. when linking newborns to deceased individuals, the tool checks whether the date of death is later than the individual's date of birth)

--ignoreBlock:(optional, default = False) add this flag without a value (i.e. True) for not requiring the equality of the last names' first letter of the matched individuals. By default, the tool only saves links between individuals that at least have the same first letter of their last names

burgerLinker v0.2

This is the first release of the burgerLinker tool. This version, released as a Runnable JAR file, supports the following functionalities:

• ShowDatasetStats
• ConvertToHDT
• Within_B_M
• Within_B_D
• Between_B_M
• Between_B_D
• Between_M_M
• Between_D_M
• Closure