# apache/predictionio

PredictionIO, a machine learning server for developers and ML engineers.

Repository: https://github.com/apache/predictionio
Canonical: https://ross.abutalabs.com/products/predictionio
Homepage: https://predictionio.apache.org
Language: Scala
License: Apache-2.0
License Family: permissive
Topics: scala, big-data, predictionio
Archived: true
Last push: 2021-01-09T10:30:58+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 4968, "days_push": 2062, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12518, forks 1897 (observed 2026-08-28T04:10:59.274346+00:00)

## What it is
Apache PredictionIO is an open-source machine learning server built on Spark, HBase, Elasticsearch, and Hadoop that lets developers build, train, and deploy predictive engines (recommendation, classification, similar-product) as REST web services. The project has been retired to the Apache Attic and is no longer maintained.

## Use cases
- deploy a recommendation engine as a REST API
- build a 'customers who viewed this also viewed' feature
- collect user events and train predictive models on Spark
- serve real-time product recommendations to anonymous visitors
- run text classification with Naive Bayes as a web service
- predict which service plan a user will subscribe to

## When to choose
- you are maintaining a legacy deployment already built on PredictionIO
- you need a self-hosted ML server stack on older Spark/HBase/Elasticsearch versions
- you want to study a Lambda Architecture ML server implementation

## When to avoid
- you are starting a new project - the project is retired and unmaintained
- you need current Spark, Elasticsearch, or Java versions
- you want modern ML tooling or active community support

## Facets
- artifact type: service
- maturity: abandoned
- function: machine-learning, api-framework, http-server, data-science, search-engine, streaming
- domain: machine-learning, big-data, data-science, e-commerce, self-hosted
- platform: jvm, self-hosted
- tags: predictionio, recommendation-engine, spark-mllib, lambda-architecture, retired-project, apache-attic, event-collection, predictive-engines, linux, docker, web-server

## Member repositories
- apache/predictionio (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:59.274346+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 2026-08-29T17:13:58.935570+00:00, confidence not recorded.
  - readme: https://github.com/apache/predictionio (fetched 2026-08-28T04:10:59.274346+00:00, sha 77ff83bdcc16)
  - homepage: https://predictionio.apache.org (fetched 2026-08-29T08:10:39.385246+00:00, sha e29b4948a4a1)
  - site_page: https://predictionio.apache.org/templates/similarproduct/quickstart (fetched 2026-08-29T08:10:39.401036+00:00, sha 87320b080961)
  - site_page: https://predictionio.apache.org/templates/classification/quickstart (fetched 2026-08-29T08:10:39.404161+00:00, sha 1eb9a6bf5e42)
  - site_page: https://predictionio.apache.org/community/contribute-documentation (fetched 2026-08-29T08:10:39.406946+00:00, sha 8bd8c66bb93c)
  - site_page: https://predictionio.apache.org/install (fetched 2026-08-29T08:10:39.394405+00:00, sha 41eb94fc82a1)
  - site_page: https://predictionio.apache.org/templates/recommendation/quickstart (fetched 2026-08-29T08:10:39.396289+00:00, sha 2c21f79c3d9f)
  - site_page: https://predictionio.apache.org/templates/ecommercerecommendation/quickstart (fetched 2026-08-29T08:10:39.398659+00:00, sha 775ea31cc813)
  - site_page: https://predictionio.apache.org/resources/faq (fetched 2026-08-29T08:10:39.409236+00:00, sha c06cfeb0e284)
- Data as of 2026-08-30T08:39:29.467469+00:00.