A Cross-platform Benchmark Framework For Mobile Semantic Web Reasoning Engines In Clinical Decision Support Systems
Notice bibliographique
Résumé
Background & Objectives Semantic Web technologies are used extensively in the health domain to enable expressive, standards-based reasoning. Deploying Semantic Web reasoning processes directly on mobile devices has a number of advantages, including robustness to connectivity loss and more timely results. By leveraging local reasoning processes, Clinical Decision Support Systems (CDSS) can thus present timely alerts given dangerous health issues, even when connectivity is lacking. However, a number of challenges arise as well, related to mobile platform heterogeneity and limited computing resources. To tackle these challenges, developers should be empowered to benchmark mobile reasoning performance across different mobile platforms, with rule- and datasets of varying scale and complexity, and under typical CDSS reasoning process flows. To deal with the current heterogeneity of rule formats, a uniform interface on top of mobile reasoning engines also needs to be provided. System We present a mobile, cross-platform benchmark framework, comprising two main components: 1) a generic Semantic Web layer, supplying a uniform, standards-based rule- and dataset interface to mobile reasoning engines; and 2) a Benchmark Engine, to investigate mobile reasoning performance. This framework was implemented using the PhoneGap cross-platform development tool, allowing it to be deployed on a range of mobile platforms. During benchmark execution, the benchmark rule- and dataset (encoded using the SPARQL Inferencing Notation (SPIN) and Resource Description Framework (RDF)) are first passed to the generic Semantic Web layer. In this layer, the local Proxy component contacts an external Conversion Web Service, where converters perform conversion into the different rule engine formats. Developers may develop new converters to support other engines. The results are then communicated back to the Proxy and passed on to the local Benchmark Engine. In the Benchmark Engine, reasoning can be conducted using different process flows, to better align the benchmarks with real-world CDSS. To plugin new reasoning engines (JavaScript or native), developers need to implement a plugin realizing a uniform interface (e.g., load data, execute rules). New process flows can also be supplied. In the benchmarks, data and rule loading times, as well as reasoning times, are measured. From our work in clinical decision support, we identified two useful reasoning process flows: * Frequent Reasoning: To infer new facts, the reasoning engine is loaded with the entire datastore each time a certain timespan has elapsed, and the relevant ruleset is executed. * Incremental Reasoning: In this case, the datastore is kept in-memory, whereby reasoning is applied each time a new fact has been added. Currently, 4 reasoning engines (and their custom formats) are supported, including RDFQuery (https://code.google.com/p/rdfquery/wiki/RdfPlugin), RDFStore-JS (http://github.com/antoniogarrote/rdfstore-js), Nools (https://github.com/C2FO/nools) and AndroJena (http://code.google.com/p/androjena/). Conclusion In this paper, we introduced a mobile, cross-platform and extensible benchmark framework for comparing mobile Semantic Web reasoning performance. Future work consists of investigating techniques to optimize mobile reasoning processes.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction distillée sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Apprise à partir de 10 348 étiquettes directes de Codex et de 10 348 étiquettes directes de Gemma. Le mode candidate est l'union des têtes enseignantes seuillées; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont ni des étiquettes humaines ni des étiquettes directes de modèles de pointe.
Scores Codex et Gemma par catégorie
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,013 | 0,015 |
| Méta-épidémiologie (sens strict) | 0,000 | 0,000 |
| Méta-épidémiologie (sens large) | 0,001 | 0,000 |
| Bibliométrie | 0,001 | 0,001 |
| Études des sciences et des technologies | 0,001 | 0,000 |
| Communication savante | 0,002 | 0,003 |
| Science ouverte | 0,002 | 0,001 |
| Intégrité de la recherche | 0,000 | 0,001 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,000 | 0,001 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule tête enseignante, pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».