Contexts in source publication

Context 1
... AnnoJob system represents a semantic annotation-based approach for job recommendation based on two main modules (detailed in Figure 1). ...
Context 2
... we implicitly search for the required information in the acquired information using semantic relatedness. Figure 10. Example of finding relatedness between an annotated offer and resume using ontology and Wikidata. ...
Context 3
... of finding relatedness between an annotated offer and resume using ontology and Wikidata. Figure 11 illustrates an example of a given offer and resume annotated with the domain ontology. From the offer, the entity "mobile app developer" is annotated with the concept "job". ...
Context 4
... the triple subjects are the same or related, we compare the objects. Figure 10 represents a required triple "having 2 years' experience in java". To match this triple, we search for triples with the same predicate. ...
Context 5
... we compare the required and the acquired diploma instances employing the relation (related to) from the ontology. Next, to match the required and the acquired degree, such as bachelor's and master's, in the ontology, we assign the annotation "isDefinedBy" with the attribute "Value" to define an integer value for each degree to check whether a candidate is underqualified compared to the academic qualification requirement (see Figure 12). (e.g., a master's degree is assigned a value of 3, while a bachelor's degree is assigned a value of 1). ...
Context 6
... implementation uses Gate embedded, Protégé OWL API to work with the OWL ontologies, and the Jena API to manipulate the different constituents of the ontology (classes, relations, instances, data type properties, etc.). Figure 13 illustrates the AnnoJob component diagram, including the two main components (information extraction and semantic matching). The aim is to upgrade, maintain, and enhance each component while keeping the overall system code unaffected. ...
Context 7
... on our experiments and studies, the similarity score threshold is defined as 0.5; If the similarity score between two entities is higher than 0.5, the two entities are correctly classified as related. As shown in Figure 17, we calculated the true positive rate (TPR; sensitivity) and the false positive rate (FPR) using different threshold options. The TPR represents what percentage we classified correctly and FPR represents what percentage we miss-classified based on the given threshold. ...

Similar publications

Preprint
Full-text available
We present an operational semantics for the language MeTTa.
Article
Full-text available
Zusammenfassung Im Rahmen dieser Studie wird gezeigt, dass Präpositionalphrasen mit duruh im Ahd. Tatian sowie in Otfrids Evangelienbuch als Ergänzung sowie als Adverbiale auftreten können. Während sich Verwendungen als Ergänzung im Wesentlichen auf lokale Semantik beschränken, weist der Bereich der Adverbialen ein großes semantisches Spektrum auf....