Analysis of the results of the paper Traveling with a Map: Optimizing the Search Space of Link Traversal Queries Using RDF Data Shapes
Python script and Jupyter notebooks for analyzing the experiment results for the paper Traveling with a Map: Optimizing the Search Space of Link Traversal Queries Using RDF Data Shapes.
The figures and the table from the paper are generated from this repository in ./artefact
.
The ./results
folder contains the raw data of the experiment.
./artefact/http_req_exec_time_relation
contains plots of the relation between HTTP request and the query execution time compared to the state-of-the-art approach, type index with LDP.
The file ./stats.json
presents some statistics about the curve of the plot.
./artefact/query_containment_execution_time
contains tables of the execution time of the query-shape containment algorithm in relation to different sets of shapes.
./artefact/ratio_useful_resources
contains a table of the ratio of useful sources of the shape index approach and the state-of-the-art.
./statistical_significance
contains tables of the statistical significance of the difference in query execution time of different experiments.
./variation_approach
contains plots of the reduction of the number of HTTP and query execution times of the shape index and LDP approach in relation to the state-of-the-art type index with LDP.
./variation_detail_shape
contains plots of the reduction of the number of HTTP and query execution time of the shape index with different levels of detail of shapes.
./variation_percentage_shape_index
contains plots of the reduction of the number of HTTP and query execution times of the shape index in networks with different percentages of datasets having a shape index.
./variation_percentage_entry_shape_index
contains plots of the reduction of the number of HTTP and query execution time of the shape index in networks with different percentages of shape index entries having closed shapes as keys.
./variation_percentage_entry_shape_index
contains a combined plot of the three previous plots.
./create_python_env.sh
source ./venv/bin/activate
To launch the jupyter notebook with the dependencies required to execute the analysis
./execute_jupyter.sh
The notebooks are located in ./notebooks
To generate all the artifacts run
./generate_artefact.sh