Color Words in Text

Misidentifying parts of speech leads to flawed sentence structure, weak writing, and failed grammar tests. Manual annotation is slow and error-prone past a few sentences. This tool applies automatic lexical classification to every word in your input using a heuristic dictionary of 1,500+ English words mapped to syntactic categories: nouns, verbs, adjectives, adverbs, prepositions, pronouns, conjunctions, articles, and numerals. Each category receives a distinct color, producing an instant visual parse of your text. The classification operates on isolated token matching. It does not resolve contextual ambiguity (e.g., "run" as noun vs. verb). Accuracy is approximately 75 - 85% on standard prose. Proper NLP parsers reach 97% but require server-side models.

The tool tokenizes input text using a regular expression pattern that separates words from whitespace and punctuation:

Each token t is classified by dictionary lookup:

Where D is a dictionary mapping lowercase word forms to part-of-speech categories. If no match exists, suffix heuristics apply:

Where t = input token, D = part-of-speech dictionary containing 1,500+ entries. The suffix heuristic layer catches inflected forms not in the base dictionary. Statistical accuracy on standard English prose: 75 - 85%.