TEXT ANALYSIS KNOWLEDGE BASE
Formulas, examples, and honest limits
Understand the calculation behind each tool before interpreting its output. Every guide documents a specific method, shows the denominator or vector model, and links back to a working analyzer.
EIGHT GUIDES
Choose the method you need to understand
Start with counts and phrase windows, then move into vector weighting, similarity, and multi-document comparison.
How to Calculate Word Frequency
Counts, percentages, rates per 1,000, tokenization choices, and a complete worked example.
Read guide →Keyword AnalysisKeyword Density Formula
Exact phrase calculations, competing denominator conventions, comparison rules, and limitations.
Read guide →NLP FundamentalsBag of Words Model
How token counts become sparse document vectors and transparent machine-learning features.
Read guide →Model ComparisonBag of Words vs Word2Vec
When transparent frequency vectors or learned word embeddings better match the task.
Read guide →Phrase AnalysisWhat Are N-grams?
Unigrams through longer phrase windows, denominator rules, filtering behavior, and examples.
Read guide →NLP FundamentalsTF-IDF Formula
Normalized term frequency, smoothed inverse document frequency, corpus effects, and a worked calculation.
Read guide →Document SimilarityCosine Similarity for Text
How BoW and TF-IDF vectors become an overlap score, with contribution terms and interpretation limits.
Read guide →Revision WorkflowCompare Texts by Word Frequency
Measure normalized vocabulary and phrase changes without confusing the result with a character diff.
Read guide →PUT THE METHOD TO WORK