PHPackages coral-media/php-ir - PHPackages - PHPackages [Skip to content](#main-content)[PHPackages](/)[Directory](/)[Categories](/categories)[Trending](/trending)[Leaderboard](/leaderboard)[Changelog](/changelog)[Analyze](/analyze)[Collections](/collections)[Log in](/login)[Sign up](/register) 1. [Directory](/) 2. / 3. [Search & Filtering](/categories/search) 4. / 5. coral-media/php-ir Abandoned → [coral-media/ext-ir](/?search=coral-media%2Fext-ir)ArchivedLibrary[Search & Filtering](/categories/search) coral-media/php-ir ================== Information Retrieval algorithms (vector space, similarity, clustering) v0.7.2(4mo ago)08MITPHPPHP >=8.2 Since Dec 18Pushed 4mo agoCompare [ Source](https://github.com/coral-media/php-ir)[ Packagist](https://packagist.org/packages/coral-media/php-ir)[ RSS](/packages/coral-media-php-ir/feed)WikiDiscussions main Synced 1mo ago READMEChangelogDependencies (5)Versions (20)Used By (0) PHP-IR ====== [](#php-ir) [![PHP](https://camo.githubusercontent.com/5a4f359a9d75caa6a221ef60f3f354e97287bef4051283935a7710e2cc7e50f1/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f7068702d382e322532422d3737374242343f6c6f676f3d706870)](https://camo.githubusercontent.com/5a4f359a9d75caa6a221ef60f3f354e97287bef4051283935a7710e2cc7e50f1/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f7068702d382e322532422d3737374242343f6c6f676f3d706870)[![License](https://camo.githubusercontent.com/e7dac58613f2dcadc20e7cd1d29effb32da6b2787ff83b9ed120f86b920bd8ff/68747470733a2f2f696d672e736869656c64732e696f2f7061636b61676973742f6c2f636f72616c2d6d656469612f7068702d6972)](https://camo.githubusercontent.com/e7dac58613f2dcadc20e7cd1d29effb32da6b2787ff83b9ed120f86b920bd8ff/68747470733a2f2f696d672e736869656c64732e696f2f7061636b61676973742f6c2f636f72616c2d6d656469612f7068702d6972) [![PHPStan](https://camo.githubusercontent.com/13f601b8b984aa928494c759cc8a1e0c7cb3edfef605b044d2d3a659b66e0482/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f7068707374616e2d6c6576656c25323031302d627269676874677265656e)](https://camo.githubusercontent.com/13f601b8b984aa928494c759cc8a1e0c7cb3edfef605b044d2d3a659b66e0482/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f7068707374616e2d6c6576656c25323031302d627269676874677265656e)[![PHPMD](https://camo.githubusercontent.com/c2321a49e1d204f5366f1de678700e2841cd5be1c726a4d31c4673ac75e33282/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f7068706d642d656e61626c65642d626c7565)](https://camo.githubusercontent.com/c2321a49e1d204f5366f1de678700e2841cd5be1c726a4d31c4673ac75e33282/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f7068706d642d656e61626c65642d626c7565) [![GitHub last commit](https://camo.githubusercontent.com/bca4f65dd886fe4a80a9b12e1c57dcc64db0db8b2e21302ad4e5d930d6aaa077/68747470733a2f2f696d672e736869656c64732e696f2f6769746875622f6c6173742d636f6d6d69742f636f72616c2d6d656469612f7068702d6972)](https://camo.githubusercontent.com/bca4f65dd886fe4a80a9b12e1c57dcc64db0db8b2e21302ad4e5d930d6aaa077/68747470733a2f2f696d672e736869656c64732e696f2f6769746875622f6c6173742d636f6d6d69742f636f72616c2d6d656469612f7068702d6972)[![GitHub repo size](https://camo.githubusercontent.com/8b5e22ac4e928b08a2d20428d1b885bbbbaf852c18243d52809c858b24fc3827/68747470733a2f2f696d672e736869656c64732e696f2f6769746875622f7265706f2d73697a652f636f72616c2d6d656469612f7068702d6972)](https://camo.githubusercontent.com/8b5e22ac4e928b08a2d20428d1b885bbbbaf852c18243d52809c858b24fc3827/68747470733a2f2f696d672e736869656c64732e696f2f6769746875622f7265706f2d73697a652f636f72616c2d6d656469612f7068702d6972) **PHP-IR** is a modern, research-oriented **Information Retrieval (IR)** and **Vector Space Modeling** library for PHP, focused on correctness, transparency, and theoretical grounding. It provides low-level, composable primitives for **text representation, weighting, similarity, clustering, and evaluation**, designed for engineers who need **full control and explainability**, not opaque ML abstractions. --- Why PHP-IR exists ----------------- [](#why-php-ir-exists) The PHP ecosystem has historically lacked serious IR tooling beyond thin wrappers around search engines. PHP-IR fills that gap by offering: - Explicit **vector space modeling** - Reproducible **term weighting pipelines** - Deterministic **clustering algorithms** - Quantitative **cluster quality metrics** - APIs aligned with **Information Retrieval literature** The goal is not convenience-first APIs, but **scientifically correct and inspectable IR workflows**. --- Core capabilities ----------------- [](#core-capabilities) ### Text processing [](#text-processing) - Tokenization (regex, whitespace) - Text normalization (lowercasing, accent folding, composition) - Stop-word filtering with language support (English, Spanish) ### Vocabulary & statistics [](#vocabulary--statistics) - Vocabulary construction - Document frequency tracking - IDF computation (per-term and vectorized) - Corpus-level statistics via dedicated façades (no core pollution) ### Vectorization [](#vectorization) - Sparse and dense vector representations - Term Frequency (TF) - TF-IDF weighting - Spherical (L2-normalized) vector spaces - Explicit densification for algorithms that require fixed dimensions ### Similarity [](#similarity) - Cosine similarity - Pluggable similarity interfaces ### Clustering [](#clustering) - Spherical K-Means - **Spherical K-Medians** (robust to outliers) - Deterministic centroid update strategies - Explicit iteration control - Centroid initialization and update policies ### Cluster evaluation [](#cluster-evaluation) - Intra-cluster cohesion - Inter-cluster separation - Global quality score aligned with IR theory - Metrics designed for **algorithm comparison**, not just reporting --- Design philosophy ----------------- [](#design-philosophy) PHP-IR is intentionally **not**: - A search engine - A machine learning framework - A black-box clustering toolkit Instead, it provides **clear, inspectable building blocks** that let you: - Reason about every step of the IR pipeline - Swap strategies without side effects - Validate theoretical assumptions with executable code - Compare algorithms using quantitative invariants If you are familiar with TF-IDF, cosine similarity, and clustering theory, PHP-IR should feel predictable and rigorous. --- Theoretical foundation ---------------------- [](#theoretical-foundation) The library is grounded in classical and modern IR research, including: - *[Introduction to Information Retrieval](https://nlp.stanford.edu/IR-book/)* - Christopher D. Manning, Prabhakar Raghavan, Hinrich Schütze - *[Spherical k-means clustering](http://www.cs.utexas.edu/~inderjit/public_papers/concept_mlj.pdf)* - I. S. Dhillon and D. S. Modha - *[Spherical K - Medians](https://www.academia.edu/27956871/Spherical_K_Medians)* - Rafael E. Espinosa Santiesteban --- Current status -------------- [](#current-status) - Actively developed - API stabilized through real-world usage - Strong test coverage with **invariant-based tests** - English and Spanish corpora used for validation - Designed to evolve without breaking theoretical guarantees > Detailed documentation, examples, and usage guides will be added incrementally. --- Roadmap (high level) -------------------- [](#roadmap-high-level) - Advanced convergence criteria beyond fixed iteration limits - Additional robustness heuristics for clustering - Optional serialization of evaluation artifacts - Extended language tooling and corpora support --- License ------- [](#license) MIT License. Use it, extend it, and build on it responsibly. ### Health Score 35 — LowBetter than 80% of packages Maintenance78 Regular maintenance activity Popularity4 Limited adoption so far Community6 Small or concentrated contributor base Maturity46 Maturing project, gaining track record Bus Factor1 Top contributor holds 100% of commits — single point of failure How is this calculated?**Maintenance (25%)** — Last commit recency, latest release date, and issue-to-star ratio. Uses a 2-year decay window. **Popularity (30%)** — Total and monthly downloads, GitHub stars, and forks. Logarithmic scaling prevents top-heavy scores. **Community (15%)** — Contributors, dependents, forks, watchers, and maintainers. Measures real ecosystem engagement. **Maturity (30%)** — Project age, version count, PHP version support, and release stability. ### Release Activity Cadence Every ~1 days Total 19 Last Release 122d ago ### Community Maintainers ![](https://avatars.githubusercontent.com/u/5040695?v=4)[Rafael Ernesto Espinosa Santiesteban](/maintainers/rernesto)[@rernesto](https://github.com/rernesto) --- Top Contributors [![rernesto](https://avatars.githubusercontent.com/u/5040695?v=4)](https://github.com/rernesto "rernesto (94 commits)") --- Tags clustering-algorithminformation-retrievalk-meansk-means-clusteringtext-classificationsearchrecommendationclusteringk-meansTF-IDFinformation-retrievaltext-similaritycosine-similarityvector-space-model ### Code Quality TestsPHPUnit Static AnalysisPHPStan Code StylePHP CS Fixer Type Coverage Yes ### Embed Badge ![Health badge](/badges/coral-media-php-ir/health.svg) ``` [![Health](https://phpackages.com/badges/coral-media-php-ir/health.svg)](https://phpackages.com/packages/coral-media-php-ir) ``` ### Alternatives [elasticsearch/elasticsearch PHP Client for Elasticsearch 5.3k178.3M943](/packages/elasticsearch-elasticsearch)[rubix/ml A high-level machine learning and deep learning library for the PHP language. 2.2k1.4M28](/packages/rubix-ml)[ruflin/elastica Elasticsearch Client 2.3k50.4M203](/packages/ruflin-elastica)[solarium/solarium PHP Solr client 93432.7M98](/packages/solarium-solarium)[opensearch-project/opensearch-php PHP Client for OpenSearch 15224.3M65](/packages/opensearch-project-opensearch-php) PHPackages © 2026 [Directory](/)[Categories](/categories)[Trending](/trending)[Changelog](/changelog)[Analyze](/analyze)