Concepts

TODO:

  1. Ontology
  2. Flat NDJSON files, no database, etc.
  3. Data (TGN, GeoNames, BAG, NWB, Who’s on First)
  4. Import API (UNIX Diff)
  5. Graphmalizer
  6. Graph - hierarchies, equivalence, concepts
  7. Search API
  8. Viewer

Data model

TODO: PITs, relations, concepts.

API

TODO: Search, geocoding. See GitHub.

Architecture

Histograph PITs Relations Meta Dataset 1 D2 D3 D4 Neo4j Plugin Elasticsearch Histograph Core Histograph API Histograph Viewer Text, temporal, and geospatial indexes. JSON geocoding API. Accepts search queries and returns GeoJSON FeatureCollections where one Feature represents one place concept. Shows place concepts and PITs on map and in a graph. Expands PITs to concepts by doing breadth-first graph searches along equivalence rela-tions. Neo4j Graph database, con-taining PITs, relations between PITs, and equiv-alence classes. Graphmalizer Manages graph of PITs and relations, independ-ent of order of messag-es on queue. Creates equivalence classes for PITs in the same con-cept. Reads message queue, transforms and distributes messages to Graphmalizer and Elasticsearch. Histograph Import Imports directory structure containing multiple datasets Redis Queue containing atomic actions, e.g. add PIT, remove relation, etc... Histograph IO Import and export of PITs and relations via HTTP. IO accepts NDJSON files, parses, checks and diffs them, and adds actions to queue. GitHub repository: histograph/import histograph/data histograph/concordances histograph/io histograph/core histograph/api histograph/viewer graphmalizer/graphmalizer-core histograph/neo4j-plugin