# WP21 — Search audit (generated by scripts/screening/search_audit.py)

Date of searches: 2026-10-20. Queries executed: 67.

## Queries and hits

| date | source | mode | query | hits | retrieved |
|---|---|---|---|---|---|
| 2026-10-20 | crossref | bibliographic | hedonic housing prices | 192459 | 4000 |
| 2026-10-20 | crossref | bibliographic | hedonic house price model | 3449996 | 4000 |
| 2026-10-20 | crossref | bibliographic | hedonic property valuation residential | 369813 | 4000 |
| 2026-10-20 | crossref | bibliographic | residential property price determinants regression | 989195 | 4000 |
| 2026-10-20 | crossref | bibliographic | implicit prices housing attributes | 296913 | 4000 |
| 2026-10-20 | crossref | bibliographic | spatial hedonic housing | 666198 | 4000 |
| 2026-10-20 | crossref | bibliographic | house prices proximity hedonic | 244545 | 4000 |
| 2026-10-20 | crossref | bibliographic | apartment rent hedonic | 24968 | 4000 |
| 2026-10-20 | crossref | bibliographic | housing price machine learning valuation | 3354684 | 4000 |
| 2026-10-20 | crossref | bibliographic | automated valuation model residential | 3403009 | 4000 |
| 2026-10-20 | crossref | bibliographic | prix hédoniques logement | 14416 | 4000 |
| 2026-10-20 | crossref | bibliographic | precios hedónicos vivienda | 6767 | 4000 |
| 2026-10-20 | crossref | bibliographic | hedonische Preise Wohnungen | 1718 | 1718 |
| 2026-10-20 | crossref | bibliographic | preços hedônicos imóveis | 1871 | 1871 |
| 2026-10-20 | semanticscholar | bulk | hedonic + (housing | house | houses | residential | apartment | apartments | condominium | | 1853 | 1853 |
| 2026-10-20 | semanticscholar | bulk | ("implicit price" | "implicit prices") + (housing | house | residential | property) | 414 | 414 |
| 2026-10-20 | semanticscholar | bulk | ("house prices" | "housing prices" | "home prices" | "property prices" | "property values" | 13866 | 13865 |
| 2026-10-20 | semanticscholar | bulk | ("automated valuation" | "mass appraisal") + (residential | housing | house | property) | 607 | 607 |
| 2026-10-20 | semanticscholar | bulk | ("machine learning" | "deep learning" | "neural network" | "random forest" | "gradient boo | 1802 | 1802 |
| 2026-10-20 | semanticscholar | bulk | ("computer vision" | "street view" | "satellite imagery" | "images") + ("house price" | "h | 2541 | 2540 |
| 2026-10-20 | semanticscholar | bulk | ("text mining" | "natural language processing" | "textual" | "listing descriptions") + ("h | 387 | 387 |
| 2026-10-20 | semanticscholar | bulk | ("spatial hedonic" | "spatial econometric" | "spatial lag" | "spatial error" | "geographic | 2889 | 2889 |
| 2026-10-20 | semanticscholar | bulk | ("meta-analysis" | "meta analysis" | "meta-regression") + (hedonic | "house prices" | "hou | 315 | 315 |
| 2026-10-20 | semanticscholar | bulk | ("willingness to pay" | capitalization | capitalisation) + hedonic + (housing | house | re | 822 | 821 |
| 2026-10-20 | semanticscholar | bulk | hedonic housing prices | 3364 | 3364 |
| 2026-10-20 | semanticscholar | bulk | hedonic house price | 2194 | 2194 |
| 2026-10-20 | semanticscholar | bulk | hedonic property valuation | 667 | 667 |
| 2026-10-20 | semanticscholar | bulk | residential property price determinants | 372 | 372 |
| 2026-10-20 | semanticscholar | bulk | housing characteristics prices hedonic | 988 | 988 |
| 2026-10-20 | semanticscholar | bulk | implicit prices housing attributes | 134 | 134 |
| 2026-10-20 | semanticscholar | bulk | real estate hedonic regression | 594 | 594 |
| 2026-10-20 | semanticscholar | bulk | spatial hedonic model housing | 952 | 952 |
| 2026-10-20 | semanticscholar | bulk | housing price meta-analysis hedonic | 23 | 23 |
| 2026-10-20 | semanticscholar | bulk | automated valuation model residential | 107 | 107 |
| 2026-10-20 | semanticscholar | bulk | text real estate valuation machine learning | 9 | 9 |
| 2026-10-20 | semanticscholar | bulk | computer vision property valuation | 24 | 24 |
| 2026-10-20 | semanticscholar | bulk | machine learning hedonic pricing housing | 74 | 74 |
| 2026-10-20 | semanticscholar | bulk | apartment prices hedonic | 455 | 455 |
| 2026-10-20 | semanticscholar | bulk | rent hedonic regression housing | 122 | 122 |
| 2026-10-20 | semanticscholar | bulk | willingness to pay housing attributes hedonic | 99 | 99 |
| 2026-10-20 | semanticscholar | bulk | house prices proximity hedonic | 263 | 263 |
| 2026-10-20 | semanticscholar | bulk | hedonic price model amenities residential | 191 | 191 |
| 2026-10-20 | semanticscholar | bulk | prix hédoniques logement | 40 | 40 |
| 2026-10-20 | semanticscholar | bulk | prix hédonique immobilier résidentiel | 3 | 3 |
| 2026-10-20 | semanticscholar | bulk | precios hedónicos vivienda | 138 | 138 |
| 2026-10-20 | semanticscholar | bulk | modelo de precios hedónicos vivienda | 96 | 96 |
| 2026-10-20 | semanticscholar | bulk | hedonische Preise Wohnimmobilien | 1 | 1 |
| 2026-10-20 | semanticscholar | bulk | hedonisches Preismodell Wohnungen | 2 | 2 |
| 2026-10-20 | semanticscholar | bulk | preços hedônicos imóveis residenciais | 19 | 19 |
| 2026-10-20 | semanticscholar | bulk | prezzi edonici abitazioni | 5 | 5 |
| 2026-10-20 | semanticscholar | bulk | 特征价格 住房 | 39 | 39 |
| 2026-10-20 | semanticscholar | bulk | 房价 特征价格模型 | 37 | 37 |
| 2026-10-20 | semanticscholar | bulk | ヘドニック 住宅価格 | 21 | 21 |
| 2026-10-20 | semanticscholar | bulk | 헤도닉 주택가격 | 10 | 10 |
| 2026-10-20 | semanticscholar | bulk | гедонистическая модель цены жилья | 0 | 0 |
| 2026-10-20 | semanticscholar | bulk | hedonik fiyat konut | 49 | 49 |
| 2026-10-20 | semanticscholar | bulk | hedoniczny model cen mieszkań | 0 | 0 |
| 2026-10-20 | semanticscholar | bulk | hedonische prijzen woningen | 0 | 0 |
| 2026-10-20 | semanticscholar | citations | cites:sirmans2005 (10.1080/10835547.2005.12090154) | nan | 0 |
| 2026-10-20 | semanticscholar | citations | cites:sirmans2006 (10.1007/s11146-006-9983-5) | nan | 237 |
| 2026-10-20 | arxiv | all | hedonic house price | 29054 | 2000 |
| 2026-10-20 | arxiv | all | house price prediction machine learning | 801055 | 2000 |
| 2026-10-20 | arxiv | all | real estate valuation deep learning | 755968 | 2000 |
| 2026-10-20 | arxiv | all | property valuation computer vision | 1519101 | 2000 |
| 2026-10-20 | arxiv | all | housing price text listing | 221354 | 2000 |
| 2026-10-20 | arxiv | all | automated valuation model | ERROR | <HTTPError 429: 'Unknown Error'> |
| 2026-10-20 | arxiv | all | hedonic regression housing | 52311 | 2000 |

## Record counts

```json
{
  "records_identified_raw": 103011,
  "by_source": {
    "crossref": 51589,
    "semanticscholar_bulk": 37585,
    "arxiv": 13400,
    "semanticscholar_citations": 237,
    "openalex": 200
  },
  "records_after_dedup": 56727,
  "records_with_abstract": 17863,
  "records_screened_rules": 56727,
  "excluded_by_rules": 49652,
  "stage1_candidates": 7075,
  "screened_by_llm": 7083,
  "llm_excluded": 3440,
  "llm_exclusion_reasons": {
    "E8_not_housing_market": 1288,
    "E2_no_characteristics": 1040,
    "E1_no_residential_price": 429,
    "E5_not_empirical_regression": 279,
    "E3_theory_method_only": 242,
    "E4_review_or_meta": 142,
    "E6_duplicate_version": 20
  },
  "llm_unclear": 428,
  "included_hedonic_stage1": 2249,
  "included_ml_predictive_stage1": 897,
  "fulltext_sought_hedonic_or_unclear": 2667,
  "fulltext_retrieved": 501,
  "fulltext_not_retrieved": 2166,
  "retrieval_attempts": 2748,
  "retrieval_status": {
    "error": 1546,
    "ok": 530,
    "not_pdf": 511,
    "no_oa_url": 161
  },
  "fulltext_assessed_extracted": 297,
  "fulltext_eligible": 242,
  "fulltext_ineligible": 55
}
```

## Coverage limitations

- OpenAlex: without an API key the service enforces a shared daily budget per IP (HTTP 401 after about 200 records on the search day); it was used only for verification.
- Semantic Scholar: unauthenticated bulk endpoint, frequent HTTP 429; all queries completed with backoff; hit counts are the service's totals, retrieval capped at 20,000 per query (never reached).
- Crossref: relevance-ranked bibliographic search returns hundreds of thousands of fuzzy matches; the first 4,000 per query were kept; most Crossref records carry no abstract, so their stage-1 screening is title-based.
- arXiv: one query failed with HTTP 429 (logged); arXiv's all-field search is fuzzy and most hits are off-topic.
- SSRN, Google Scholar, IDEAS/EconPapers, CORE, DOAJ, institutional repositories: no compliant bulk access from this environment; grey literature is covered only indirectly through Semantic Scholar and arXiv indexing.
- Full texts: only legal open-access copies (Unpaywall locations, Semantic Scholar OA links, arXiv); paywalled studies that passed stage 1 are counted as 'not retrieved' in the PRISMA flow.