Skip to main content

​
GET /api/search

Search through cached Kaggle competitions using a hybrid search strategy that combines:
  • Exact slug matching for precise lookups
  • Full-text search (FTS) using PostgreSQL’s tsvector for natural language queries
  • Fuzzy matching with trigram similarity (pg_trgm) for typo-tolerant searches
Only competitions with status = 'completed' are included in search results.

​
Endpoint

GET /api/search

​
Query Parameters

​
query
string
required
Search query string. Can be:
  • Exact slug: titanic
  • Partial title: house prices
  • Keywords: regression, nlp, computer vision
  • Fuzzy match: titanik (will match “titanic”)
Searches across competition slugs, titles, and descriptions.

​
Response

​
results
array
Array of matching competition objects, ordered by relevance score (highest first)

​
Search Strategy

The search uses a multi-strategy approach with three parallel queries:

​
1. Exact Match (Score: 10.0)

Matches the exact competition slug: 
WHERE slug = 'titanic' AND status = 'completed'

​
2. Full-Text Search (FTS)

Uses PostgreSQL’s built-in FTS with ts_rank for relevance scoring: 
WHERE search_vector @@ plainto_tsquery('english', 'house prices')
Searches across:
  • Competition title
  • Description
  • Metadata fields

​
3. Fuzzy Trigram Matching

Handles typos and partial matches using pg_trgm: 
WHERE similarity(title, 'titanik') > 0.1
Examples:
  • titanik → matches titanic
  • regresion → matches regression
  • house pric → matches house prices

​
Result Ranking

Results are deduplicated and sorted by the highest score from any matching strategy:
  1. Exact slug matches appear first (score: 10.0)
  2. Followed by FTS matches (ranked by relevance)
  3. Then fuzzy matches (ranked by similarity)
Maximum 20 results returned per query.

​
Examples

​
Request: Exact Slug

curl "https://api.kaggleingest.com/api/search?query=titanic"

​
Response: Exact Match

{
  "results": [
    {
      "slug": "titanic",
      "title": "Titanic - Machine Learning from Disaster",
      "description": "Start here! Predict survival on the Titanic and get familiar with ML basics",
      "metadata": {
        "title": "Titanic - Machine Learning from Disaster",
        "url": "https://www.kaggle.com/c/titanic",
        "category": "Getting Started",
        "prize": "Knowledge",
        "evaluation": "Accuracy",
        "team_count": 15234,
        "leaderboard": []
      },
      "score": 10.0
    }
  ]
}

curl "https://api.kaggleingest.com/api/search?query=house+prices"

​
Response: Multiple Matches

{
  "results": [
    {
      "slug": "house-prices-advanced-regression-techniques",
      "title": "House Prices: Advanced Regression Techniques",
      "description": "Predict sales prices and practice feature engineering, RFs, and gradient boosting",
      "metadata": {
        "title": "House Prices: Advanced Regression Techniques",
        "url": "https://www.kaggle.com/c/house-prices-advanced-regression-techniques",
        "category": "Getting Started",
        "prize": "Knowledge",
        "evaluation": "RMSE",
        "team_count": 5234,
        "leaderboard": []
      },
      "score": 0.8743
    },
    {
      "slug": "california-house-prices",
      "title": "California House Prices",
      "description": "Predict house prices in California using census data",
      "metadata": {
        "title": "California House Prices",
        "url": "https://www.kaggle.com/c/california-house-prices",
        "category": "Playground",
        "prize": "Knowledge",
        "evaluation": "RMSE",
        "team_count": 1523,
        "leaderboard": []
      },
      "score": 0.6421
    }
  ]
}

​
Request: Fuzzy Search (Typo)

curl "https://api.kaggleingest.com/api/search?query=titanik"

​
Response: Fuzzy Match

{
  "results": [
    {
      "slug": "titanic",
      "title": "Titanic - Machine Learning from Disaster",
      "description": "Start here! Predict survival on the Titanic and get familiar with ML basics",
      "metadata": {
        "title": "Titanic - Machine Learning from Disaster",
        "url": "https://www.kaggle.com/c/titanic",
        "category": "Getting Started",
        "prize": "Knowledge",
        "evaluation": "Accuracy",
        "team_count": 15234,
        "leaderboard": []
      },
      "score": 0.7142
    }
  ]
}

​
Request: Empty Results

curl "https://api.kaggleingest.com/api/search?query=nonexistent"

​
Response: No Matches

{
  "results": []
}

​
Error Responses

​
429 Too Many Requests

Rate limit exceeded (20 requests per minute). 
{
  "detail": "Rate limit exceeded"
}

​
500 Internal Server Error

Search query failed due to database or internal error. 
{
  "detail": "Search failed"
}

​
Notes

  • Only cached competitions with status = 'completed' are searchable
  • Empty query returns an empty results array
  • Results are limited to 20 competitions per query
  • Trigram similarity threshold is 0.1 (10% similarity)
  • Requires PostgreSQL extensions: pg_trgm and tsvector
  • Search is case-insensitive
  • Duplicate results are automatically deduplicated by slug

Build docs developers (and LLMs) love

Get started for freeTalk to us