The most popular terms we extracted were related to community coalitions, networks or partnerships (n=20, 10.8%), community health agencies, organizations or centres (n=17, 9.1%), neighborhood associations, congregations, health centers or organizations (n=9, 4.8%), non-profit organizations, agencies, consortium or sector (n=10, 5.4%) or community agencies (n=5, 2.7%).
To obtain the semantic tags for the recommendation, we proposed a process in which, after preprocessing the Web resources, the folksonomy and the personomy terms, we extract a lightweight ontology from them and create a ranking based on the terms frequency and a mapping between the Web resource and folksonomy ontologies over the personomy ontology.
To investigate the relationship among the seed terms, we extract a subgraph connecting the seed terms, including both inherited term term and augmented term term edges [see Equations (2) and (3)], such that the sum of all edge lengths for the subgraph is minimized for all possible subgraphs connecting the terms.
For each GO term, we extracted the full "pedigree" of the term by finding all possible paths from the GO term to the root term following (i) 'is_a' and 'part_of' relationships, and (ii) by following only 'is_a' relationships.
Using 12 extreme preterm birth cases and 12 matched, term controls, we extracted DNA from archived neonatal blood spots and blood collected in a similar way at 18 years of age.
Term generation: we extract terms from English text, which we tokenize before POS-tagging, sentence identification, noun phrase and local abbreviation detection.
In terms of blinding, we extracted the literal term reported by the authors in the manuscript and classified the blinding into open-label, single, double, triple and "ambiguously stated" blinding.
Then we extracted the terms from the bios of this reference user set and compared the terms found to those in the bios of the (i) first person flu users and (ii) first person unemployment users.
In total, we extracted 2490 terms, with 4101 references to 155 of 250 pages in the publication.
In the knowledge base, concepts are represented as classes and the relations of the concepts are represented in terms of prototype instances, so we extracted all the relations of the prototype instances and directly attached them to the concepts.
Second, we extracted more terms from a rapid review of the literature on PHC in South Africa.
