1 / 9

When androids dream of … oh you know

When androids dream of … oh you know. GraphQL. UI: Display. UI: Also with editing. Authorization/ Permissions. Translation. Authorization/ Permissions. RDF Management Process: Translate UI actions into RDF adds/deletions. Search Index API. Shapes. Triple Pattern Fragments.

ama
Download Presentation

When androids dream of … oh you know

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. When androids dream of … oh you know

  2. GraphQL UI: Display UI: Also with editing Authorization/ Permissions Translation Authorization/ Permissions RDF Management Process: Translate UI actions into RDF adds/deletions Search Index API Shapes Triple Pattern Fragments Configuration: Mapping/ Events and Triggers Triplestore Search Index: Solr or ElasticSearch RDF Data Indexing Process Ingest/Update Combine Ontology RDF Service/SPQARL Authorization/ Permissions RDF Service/SPQARL Data Reasoning Process Conforms to specific patterns Configuration: Results/ Events and Triggers Shapes Asset Store

  3. Generate serialization of RDF Data in requested format: N3, JSON-LD, quads,etc. Extract/Transform: Map to common model based on shapes Validate data based on shapes. Deduplicate/disambiguate. Data Source 1 Map Validate Common Model RDF Data Data Source 2 Map Resolve/ Disambiguate Data Source 3 Additions that can be sent to VIVO using the ingest/update API Map Person Shape <person> authored <publication>… Grant Shape <grant> has investigator <person>… Publication Shape <publication> has author <person>…

  4. Persons, publications, grants, and related research information is harvested from multiple sources and made available as a change set with additions to be stored in the triple store. The ingest API (which accepts additions and retractions) populates the triple store with the provided data. Images can be added separately to the asset store using the image upload process/API. Change Set: + Persons + Publications + Grants Triplestore Validated RDF Data Ingest/Update Combine Ontology RDF Service/SPQARL Authorization/ Permissions RDF Service/SPQARL Persons Publications Grants Conforms to specific patterns Asset Store Shapes Upload images

  5. The reasoning process can be triggered independently or in conjunction with an update. A potential new feature would be to enable configuration of the reasoned to specify through configuration what triples should be inferred and added to the triple store. Event Configuration: Trigger reasoning when requested. Trigger reasoning when new triples are added. Triplestore Reasoning Process Add: Class hierarchy for type <person> rdf:typefoaf:Person Ontology Added triple: <person> rdf:typevivo:Faculty. RDF Service/SPQARL Persons Publications Grants Reasoner Configuration: Create following inferences based on added triples.

  6. The indexing process, further described on the next slide, can be configured to run when specific events occur (such as update to a particular ‘shape’ or entity type or on ingest) and/or on request. Queries define how information will be extracted and added to the search index. The “index” may actually be multiple indices, with one central general search index and multiple shape-based indices. Indexing Process Search Index Index update: Extract person, publication, grant relationships Granular/Shape update: Extract person and related info to update index Triplestore Ontology RDF Service/SPQARL Persons Publications Grants Person: <person> authored <publication> . <publication> rdfs:label “Grapes of Wrath”. <publication> output of <grant> . <grant> funded by <organization> . Publication:

  7. Indexing Process Execute Queries Add structured document Search Indexer: Document Modifiers (i.e. mapping config) Event Configuration: Trigger Indexing When Indexing Request Received. Trigger Indexing When New Triples Added using Ingest/Update API. Search Engine Interface Populate index Solr Implementation Solr Mapping Configuration (Index field => Query): Entity Type => Select ?type WHERE {?individual …} Person Position => Select ?label WHERE {?person…} Publication Author => Select ?person WHERE {?publication …} Grant Investigator => Select ?person WHERE {?grant …} Elastic Search Implementation Elastic Search Person Shape Publication Shape Grant Shape

  8. Indexing Process Execute Queries Add structured document Search Indexer: Document Modifiers (i.e. mapping config) Search Engine Interface Indexing decisions: What portion of index is updated when? Shapes could inform indexing process. E.g. If code can determine a person’s info has been updated, that shape can be used to make the appropriate updates. Populate index Solr Implementation Solr Elastic Search Implementation Elastic Search Person Shape Publication Shape Grant Shape

  9. Externalization may involve channeling requests to triple store through an API layer e.g. RDF Service Interface or SPARQL API RDF Service/SPQARL Execute SPARQL Query Create URI Submit change set: + Additions - Retractions Compare graph with serialization Output graph: single or entire graph store Triplestore Ontology Data

More Related