Naver Public Places & Data Quality Analysis

Collect public Naver Map place results for location research and data cleanup. Returned names, displayed areas and categories come from the rendered search page. Native place IDs come from the page DOM; no synthetic place IDs, phone numbers or coordinates are supplied.

Example input and quick start

Start with one result and a single page. Use the public source controls below, or supply your own JSON/dataset for evidence analysis. Clear supplied JSON to [] for public collection.

{
  "searchKeywords": [
    "Seoul coffee"
  ],
  "maxResults": 1,
  "maxResultsPerKeyword": 1,
  "maxPublicPages": 1
}

Output

Illustrative shape; these example values are not current source quotes:

{
  "placeId": "illustrative-native-place-id",
  "name": "Illustrative cafe",
  "address": "Displayed neighborhood, Seoul",
  "category": "Cafe",
  "hoursKind": "not_collected",
  "sourceUrl": "https://map.naver.com/p/search/Seoul%20coffee",
  "sourceVerified": true
}

The default dataset contains meaningful requested records. PUBLIC_COLLECTION_SUMMARY reports top-level rendered pages, source outcomes, confirmed delivery and limitations. EVIDENCE_ANALYSIS describes confirmed delivered rows only. No diagnostic placeholder is billed as a result.

Analyze your evidence

Use providedRecords or providedDatasetId to validate and deduplicate your own exports. Supplied records remain sourceVerified false. Optional analysisQuery returns literal matches with evidence references. This Actor adds: Field completeness, valid coordinate coverage for supplied coordinates, and possible duplicates sharing the same supplied name and address. Duplicate candidates require review. Zero coordinates are preserved as observations rather than silently treated as missing.

Coverage and limits

Coverage is primarily South Korea. Search pages can show a neighborhood rather than a full street address; address preserves that displayed area. Phone, website, coordinates and opening-hour schedules are not collected from search cards. Unknown fields stay absent, and hoursKind is not_collected. Direct place/detail URLs and exhaustive keyword pagination are unavailable.

One maxPublicPages allowance covers top-level pages across selected sources. Browser-owned website requests render normally; Actor code does not call source APIs or intercept their response bodies. Blocking or page changes can reduce results or cause a clear empty-collection failure. Public collection is a bounded sample, not a complete platform export. Supplied records take precedence over legacy source defaults. Existing advanced inputs marked inactive in Console do not enable source API services.

The existing result event applies only to meaningful delivered records. Check the Store pricing panel for current rates.

Python API example

import os
from apify_client import ApifyClient
client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("xmmQiVCXPScRngyjo").call(run_input={
    "searchKeywords": [
        "Seoul coffee"
    ],
    "maxResults": 1,
    "maxResultsPerKeyword": 1,
    "maxPublicPages": 1
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

Support

Open the Actor Issues tab with a run ID and a sanitized input example. The run output links include collection or validation summaries for blocked sources, invalid input and partial results.