SynthLC

doi:doi:10.57702/oyfz6rmc

SynthLC

SynthLC models lung cancer patient data. Each patient has the following attributes:

Age category
Sex
Smoking habit
Comorbidities
Biomarkers
Drugs taken
Relapse

The attribute values are assigned at random. Hence, no patterns within real lung cancer patients can be observed. Alongside a generator for synthetic lung cancer data, there are already generated datasets modeling 1,000, 10,000, and 100,000 patients, respectively. The following files are part of this SynthLC entry.

SynthLC CSV: The 1k, 10k, and 100k dataset in CSV format.
SynthLC RDF: The 1k, 10k, and 100k dataset in RDF.
SynthLC Virtuoso: The 1k, 10k, and 100k dataset preloaded in Virtuoso 07.20.3238.
SynthLC Shapes: 25 SHACL shapes consisting of biomarker, drug, and relapse combinations. There are two variants, one is using SPARQL constraints while the other uses a non-standard approach for specifying the shapes target via query.
SynthLC Generator: The script used to create the 1k, 10k, and 100k dataset, and the shapes. It can be used to create more shapes or datasets of a different size.

BibTex:

@dataset{Philipp_D_Rohde_and_Maria-Esther_Vidal_2024,
    abstract = {SynthLC models lung cancer patient data. Each patient has the following attributes:

* Age category
* Sex
* Smoking habit
* Comorbidities
* Biomarkers
* Drugs taken
* Relapse

The attribute values are assigned at random. Hence, no patterns within real lung cancer patients can be observed.
Alongside a generator for synthetic lung cancer data, there are already generated datasets modeling 1,000, 10,000, and 100,000 patients, respectively. The following files are part of this SynthLC entry.

* __SynthLC CSV__: The 1k, 10k, and 100k dataset in CSV format.
* __SynthLC RDF__: The 1k, 10k, and 100k dataset in RDF.
* __SynthLC Virtuoso__: The 1k, 10k, and 100k dataset preloaded in Virtuoso 07.20.3238.
* __SynthLC Shapes__: 25 SHACL shapes consisting of biomarker, drug, and relapse combinations. There are two variants, one is using SPARQL constraints while the other uses a non-standard approach for specifying the shapes target via query.
* __SynthLC Generator__: The script used to create the 1k, 10k, and 100k dataset, and the shapes. It can be used to create more shapes or datasets of a different size.},
    author = {Philipp D. Rohde and Maria-Esther Vidal},
    doi = {10.57702/oyfz6rmc},
    institution = {TIB},
    keyword = {'cancer', 'lung cancer', 'synthetic'},
    month = {dec},
    publisher = {TIB},
    title = {SynthLC},
    url = {https://service.tib.eu/ldmservice/dataset/synthlc},
    year = {2024}
}