AYNEXT - tools for streamlining the evaluation of link prediction techniques
AYNEXT is an open source Python suite aimed towards researchers in the field of link prediction in Knowledge Graphs. Link prediction consists of predicting missing edges in a Knowledge Graph, which usually involves the application of different techniques to generate negative examples (false triples)...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-07-01
|
| Series: | SoftwareX |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S235271102300170X |
| _version_ | 1827812149960900608 |
|---|---|
| author | Fernando Sola Daniel Ayala Rafael Ayala Inma Hernández Carlos R. Rivero David Ruiz |
| author_facet | Fernando Sola Daniel Ayala Rafael Ayala Inma Hernández Carlos R. Rivero David Ruiz |
| author_sort | Fernando Sola |
| collection | DOAJ |
| description | AYNEXT is an open source Python suite aimed towards researchers in the field of link prediction in Knowledge Graphs. Link prediction consists of predicting missing edges in a Knowledge Graph, which usually involves the application of different techniques to generate negative examples (false triples) to fit a model, and splitting edges into training, testing and validation sets. Setting up a correct evaluation setup or testing new negatives-generation strategies becomes more challenging as more complex strategies and considerations (e.g., removal of inverse relations) develop. AYNEXT makes it easy to configure and customize the creation of evaluation datasets and the computation of evaluation metrics and statistical significance tests for each pair of link prediction techniques. AYNEXT has been designed to be simple to use, but modular enough to enable customization of the main steps in the evaluation process. AYNEXT-DataGen covers the pre-processing, splitting, and negatives generation steps of the evaluation process, while AYNEXT-ResTest covers the metrics computing and the statistical tests. AYNEXT offers a simple to use command line interface that takes as input either a Knowledge Graph in standard formats or the results of applying existing techniques, but can be used programmatically for in-depth customization. |
| first_indexed | 2024-03-11T23:14:54Z |
| format | Article |
| id | doaj.art-c943affe0d2d4bc3a2d9652cc2e504c2 |
| institution | Directory Open Access Journal |
| issn | 2352-7110 |
| language | English |
| last_indexed | 2024-03-11T23:14:54Z |
| publishDate | 2023-07-01 |
| publisher | Elsevier |
| record_format | Article |
| series | SoftwareX |
| spelling | doaj.art-c943affe0d2d4bc3a2d9652cc2e504c22023-09-21T04:37:36ZengElsevierSoftwareX2352-71102023-07-0123101474AYNEXT - tools for streamlining the evaluation of link prediction techniquesFernando Sola0Daniel Ayala1Rafael Ayala2Inma Hernández3Carlos R. Rivero4David Ruiz5Universidad de Sevilla, ETSII, Avda. Reina Mercedes, s/n. Sevilla, Spain; Corresponding author.Universidad de Sevilla, ETSII, Avda. Reina Mercedes, s/n. Sevilla, SpainOkinawa Institute of Science and Technology Graduate University, 7542 Onna, Onna-Son, Kunigami, Okinawa 904-0411, JapanUniversidad de Sevilla, ETSII, Avda. Reina Mercedes, s/n. Sevilla, SpainRochester Institute of Technology, 92 Lomb Memorial Drive, Rochester, NY, USAUniversidad de Sevilla, ETSII, Avda. Reina Mercedes, s/n. Sevilla, SpainAYNEXT is an open source Python suite aimed towards researchers in the field of link prediction in Knowledge Graphs. Link prediction consists of predicting missing edges in a Knowledge Graph, which usually involves the application of different techniques to generate negative examples (false triples) to fit a model, and splitting edges into training, testing and validation sets. Setting up a correct evaluation setup or testing new negatives-generation strategies becomes more challenging as more complex strategies and considerations (e.g., removal of inverse relations) develop. AYNEXT makes it easy to configure and customize the creation of evaluation datasets and the computation of evaluation metrics and statistical significance tests for each pair of link prediction techniques. AYNEXT has been designed to be simple to use, but modular enough to enable customization of the main steps in the evaluation process. AYNEXT-DataGen covers the pre-processing, splitting, and negatives generation steps of the evaluation process, while AYNEXT-ResTest covers the metrics computing and the statistical tests. AYNEXT offers a simple to use command line interface that takes as input either a Knowledge Graph in standard formats or the results of applying existing techniques, but can be used programmatically for in-depth customization.http://www.sciencedirect.com/science/article/pii/S235271102300170XKnowledge graphsEvaluationLink prediction |
| spellingShingle | Fernando Sola Daniel Ayala Rafael Ayala Inma Hernández Carlos R. Rivero David Ruiz AYNEXT - tools for streamlining the evaluation of link prediction techniques SoftwareX Knowledge graphs Evaluation Link prediction |
| title | AYNEXT - tools for streamlining the evaluation of link prediction techniques |
| title_full | AYNEXT - tools for streamlining the evaluation of link prediction techniques |
| title_fullStr | AYNEXT - tools for streamlining the evaluation of link prediction techniques |
| title_full_unstemmed | AYNEXT - tools for streamlining the evaluation of link prediction techniques |
| title_short | AYNEXT - tools for streamlining the evaluation of link prediction techniques |
| title_sort | aynext tools for streamlining the evaluation of link prediction techniques |
| topic | Knowledge graphs Evaluation Link prediction |
| url | http://www.sciencedirect.com/science/article/pii/S235271102300170X |
| work_keys_str_mv | AT fernandosola aynexttoolsforstreamliningtheevaluationoflinkpredictiontechniques AT danielayala aynexttoolsforstreamliningtheevaluationoflinkpredictiontechniques AT rafaelayala aynexttoolsforstreamliningtheevaluationoflinkpredictiontechniques AT inmahernandez aynexttoolsforstreamliningtheevaluationoflinkpredictiontechniques AT carlosrrivero aynexttoolsforstreamliningtheevaluationoflinkpredictiontechniques AT davidruiz aynexttoolsforstreamliningtheevaluationoflinkpredictiontechniques |