Book URI: https://leanpub.com/javaai

You can read my book for free online at: https://leanpub.com/javaai/read

This example provides a lightweight NLP toolkit written from scratch in Java. It includes named entity recognition using bundled name lists, a part-of-speech tagger (FastTag), an automatic keyword/topic extractor (AutoTagger), a text tokenizer, noise-word filtering, and a Porter stemmer. No external NLP service or large model download is required.

• Java 8+
• Maven 3.6+

# Named entity recognition demo
make names

# Automatic keyword extraction demo
make autotagger

# Part-of-speech tagging demo
make fasttag

Or manually:

mvn install -DskipTests -q
mvn exec:java -Dexec.mainClass="com.markwatson.nlp.ExtractNames" -q

• ExtractNames.java — Named entity recognition using serialized name databases (test_data/propernames.ser)
• FastTag.java — Rule-based part-of-speech tagger
• AutoTagger.java — Automatic keyword/topic extractor using TF-IDF-style scoring
• ComparableDocument.java — Document similarity comparison
• util/ — Tokenizer, noise-word filter, scored lists, and other helpers

Note: The file test_data/propernames.ser is a serialized Java object containing first names, last names, and place names. On first run, the ExtractNames constructor writes these out to human-readable .txt files for inspection.

This example is released using the Apache 2 license.

Copyright 2022-2026 Mark Watson. All rights reserved.

You are free to share and adapt this content, with attribution.