Dataset Overview
| Dataset Name | Land Expropriation Compensation Project Dataset |
| Total Data Volume | 8,400 project cases (covering house demolition and relocation + the full land expropriation process) |
| Type of project | Expropriation of houses on state-owned land / collectivization of collectively owned land / urban village improvement / shantytown improvement / old town renovation |
| Time Span | 2012 - 2026 |
| CoverageDimension | Project profile, compensation plan, contract execution progress, funds management, judicial procedures, petition and correspondence security |
| Data Source | Government office systems, natural resources sector, housing sector public data, network of adjudicatory documents |
| Format Output | JSON / CSV / Excel / Knowledge Graph |
Detailed Description of Data Dimensions
| Field Name | Data Type | Value Range / Description |
|---|---|---|
| project_id | String(32) | Item _UNIQUE_TECH_Evil _ID_UNIQUE_TECH__ |
| project_name | Text | Project name (de-identified) |
| project_type | Enum | HOUSE_EXPROPRIATION | LAND_ACQUISITION | URBAN_RENEWAL | SHANTYTOWN |
| region | String | Province-City-District |
| total_area_sqm | Float | Total area acquired (sq m) |
| household_count | Integer | Number of households involved |
| total_budget | Float | Total project budget (thousands of dollars) |
| compensation_mode | Enum | MONETARY | PROPERTY_SWAP | VOUCHER | HYBRID |
| signing_rate | Float | Signing rate 0.0 - 1.0 |
| project_duration_days | Integer | Project duration (days) |
| litigation_count | Integer | Number of cases involving the judiciary |
| milestones | Array | [{phase, date, status, remark}] |
De-identified JSON Data Sample
AI Training Use Cases
Projected Compensation Amount
Models applied: XGBoost / LightGBM / CatBoost. A compensation amount projection model based on multi-dimensional training of historical project data, supporting budgeting.
Contract Signing Rate Forecasting and Risk Warning
Models applied: Gradient Boosting + SHAP. Forecasts the project signing rate, identifies the key factors affecting contract signing, and provides early risk warnings.
Project Milestone Schedule Collection
Models applied: CCM/PERT + Reinforcement Learning (DQN). A lead-time strategy model trained on historical project critical path data.
Forecast of Administrative Litigation Outcomes
Applied model: Legal-BERT + Multi-Coder. Based on the training model for training in recovery justice cases, supporting response strategy development.
Get This Dataset
The recovery project dataset supports the customization of sub-authorizations by region, type, and year.
137-5502-0164 Online ConsultationRelated Datasets
Access to datasets for expropriation decisions
Changsha Langhui Information Technology Co., Ltd. The DataAssetsAPI platform is dedicated to providing AI enterprises and scientific institutions with high-quality, compliant data assets.
Data Sample Preview · Customized Data Solutions · Professional Technical Support