Frame the situation room's first question

Import the COVID-19 training scenario and define the analysis question and asset boundaries.

Question for this chapter

Which data must be connected to name the first area to respond to with evidence?

Today's reports cover Junggye 4-dong, Seongnam-dong, and Junggye Bon-dong. The goal is not to declare a vague “danger area,” but to set the order for checking movement intersections.

Why this matters now

Memorizing an asset list first can separate the execution from its decision purpose. Fix the question: compare patients, population, clinics, and hotspots for the same area.

Try it

covid19.zip Download(71 KB)

Open Collections in the left sidebar and choose More → Import in the upper-right. Upload the ZIP. If validation finds no conflicts, select Start import and wait for completion.

Import item in a collection's More menu — Choose More → Import in the Korean Portal used for the verified run.

COVID-19 scenario ZIP validation result and Start import button — Verify the scenario name and conflict state before starting import.

Success looks like this

COVID-19 Data Collection appears in the left collection list, and its detail page contains datasets, code, and pipelines.

COVID-19 Data Collection and its imported assets — Inspect the asset groups after import completes.

Interpret the result

This Workshop's final evidence appears in Graph explorer and a query-result table. Each imported asset answers a different analysis question.

What the asset shows	Question it can answer	Later use
Population data	How do people who live here differ from people who spend time here?	Compare regional scale and activity.
Patient and hotspot data	Where did patient movements intersect with places and areas?	Narrow areas for relationship review.
Clinic data	Are clinical resources present around the area?	Explain the follow-up order.
Region hierarchy	Which sigungu and sido contain a dong observation?	Report to a higher regional unit.

Datasets hold individual facts, pipelines make them comparable, and the ontology traces relationships among the facts. The next chapters connect those roles in order.

Use this shared decision table for all three reports:

Signal	Workshop threshold	Meaning
Patient count	3 or more	Several resident patients are observed.
Hotspot count	5 or more	Many movement-intersection places need review.
Patients per 10,000 residents	10 or more	The patient signal remains high after population adjustment.

An area becomes a priority candidate when it satisfies at least two signals. If candidates tie, put the one with more hotspots first because the current task is movement review. This is a learning rule for synthetic data, not a public-health policy.

Deep dive — imported assets

Asset type	Count	Contents
Collection	1	COVID-19 Data Collection (`covid19`)
Datasets	9	Population, patient, clinic, and hotspot sources plus regional intermediate results
Code	4	Population join, sido/sigungu filtering, ontology materialization
Batch pipelines	4	Execution paths for the four code assets
Ontology	6 entities / 6 relations	Region hierarchy, patient, clinic, hotspot, and six connections
Tool	1	`ontology_graph_query`
Agent	1	`epidemic_assistant`

The ZIP includes tool and Agent definitions, but the current import surface does not register them.

Next decision

Before connecting data, align a hierarchy such as 서울특별시, 노원구, and 중계4동 so the names refer to one complete place.