# ChenglongChen/kaggle-CrowdFlower

1st Place Solution for CrowdFlower Product Search Results Relevance Competition on Kaggle.

Repository: https://github.com/ChenglongChen/kaggle-CrowdFlower
Canonical: https://ross.abutalabs.com/products/kaggle-crowdflower
Homepage: https://www.kaggle.com/c/crowdflower-search-relevance
Language: C++
License Family: other
Topics: kaggle-crowdflower, search-relevance, natural-language-processing, nlp, kaggle-competetion, relevance-competition, semantic-matching, kaggle, search-engine, product-search, crowdflower, semantic-similarity
Last push: 2021-09-25T02:32:49+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 4070, "days_push": 1804, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1770, forks 649 (observed 2026-08-28T04:05:34.011226+00:00)

## What it is
The 1st place winning solution code for the CrowdFlower Search Results Relevance competition on Kaggle. It provides a full pipeline for feature extraction, XGBoost model training, and ensemble selection for predicting search result relevance.

## Use cases
- predict search results relevance for product queries
- learn how to win a kaggle nlp competition
- build semantic matching features for query-product pairs
- ensemble multiple model submissions for better scores
- study feature engineering for text relevance tasks
- reproduce a winning xgboost linear booster solution

## When to choose
- you want to study a proven top-ranked approach to search relevance prediction
- you need a reference pipeline for text feature engineering and model ensembling in Kaggle competitions
- you are working on product search relevance or semantic similarity problems

## When to avoid
- you need a production-ready or maintained search relevance library
- you want a clean modular framework (the author recommends the Kaggle_HomeDepot repo instead)
- you need actively supported software with a license

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: nlp, search-engine, machine-learning, data-science
- domain: machine-learning, data-science
- platform: python, cpp, cli
- tags: kaggle, competition-solution, search-relevance, semantic-similarity, xgboost, feature-engineering, ensemble-learning, natural-language-processing, search

## Member repositories
- ChenglongChen/kaggle-CrowdFlower (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:34.011226+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:26:02.104180+00:00, confidence not recorded.
  - readme: https://github.com/ChenglongChen/kaggle-CrowdFlower (fetched 2026-08-28T04:05:34.011226+00:00, sha 7455d5684ca9)
  - homepage: https://www.kaggle.com/c/crowdflower-search-relevance (fetched 2026-08-29T11:04:19.701317+00:00, sha 36ce1a366e8e)
- Data as of 2026-08-30T08:39:29.467469+00:00.