Letter Frequency Analyzer

This online program performs the letter frequency analysis and displays the letter distribution of any text. Frequency analysis is a cryptanalysis technique that focuses on counting letters or groups of letters in the encrypted text to aid in breaking it. Frequency analysis assumes that in sufficiently large texts of one language, the average frequency of occurrence of a certain letter is the same. For example, in most English texts, the letter "e" appears most often (12.7%), followed by the letters "t" (9.06%), "a" (8.17%), and "o" (7.5%). The least common letters in English are "x" (0.15%), "q" (0.095%), and "z" (0.074%). In Finnish, the dominant letters are "a" (12.2%), "i" (10.8%), and "n" (8.8%), and in Dutch, major letters are "e" (18.9%), "n" (10%), and "a" (7.5%). Frequency analysis can also find distributions of pairs of letters (bigrams), triplets of letters (trigrams), and longer letter groups (n-grams). Using this statistical information, it's easy to identify the language a text is written in. For example, it's been calculated that the most repeated bigrams in the English language are "th" (1.52%), "he" (1.25%), and "in" (0.94%). You can specify the length of a letter group in the options and examine the n-gram distribution of any language. When generating bigrams, trigrams, or 4+grams, you have three choices that control how they are generated. The "Join Words Together" option will remove all spacing and special symbols between all words and then split it into n-grams. For example, if you are generating bigrams and for the input data "hi world", then this string will be joined together as "hiworld", and split into bigrams "hi, iw, wo, or, rl, ld". If the option "Stop at Word Boundary" is selected, then the n-gram generator will stop at the first non-word character. If the input is plaintext "hi world" and you're generating bigrams, then the output will be "hi, wo, or, rl, ld". When generating n-grams, all words with fewer characters than n will be dropped. With the same input "hi world" and now with trigrams, the word "hi" will be dropped, and you will get "wor, orl, rld". The last option, "Include Word Boundaries" will convert spaces and special symbols in the generated n-gram list to the "·" symbol. For example, if the input is "hi world" and bigrams are being computed, then the output will be "hi, i·, ·w, wo, or, rl, ld". If there are multiple special symbols in a row, they will be ignored and only one "·" will appear in the output. For example, trigrams with the input "hello (world)" will produce "hel, ell, llo, lo·, o·w, ·wo, wor, orl, rld, ld·". Sometimes, you need to analyze more symbols than just letters. We thought of that and added the "Analyze All Symbols" option. When it's selected, you will get the statistics of numbers, punctuation marks, and any other glyphs that are used in the text. By default, the program displays only the number of occurrences of letters (or groups of letters) but you can also display percentages or totals in the count. You can also rearrange the output of the analysis and sort the counts by frequency (highest to lowest) or alphabetically (letters from a to z). By default, the frequency analysis algorithm ignores the case of letters and assumes that the letters "A" and "a" are the same. If you require case sensitive analysis, then you can turn off the ignore-case option and count the frequency of letters "A" and "a" separately. Cryptabulous!