UNPKG

crazyhorse

Version:

This module takes sentences and remixes them into a different semi-coherent sentence.

130 lines (105 loc) 6.14 kB
crazyhorse ========== [![NPM](https://nodei.co/npm/crazyhorse.png?downloads=true&downloadRank=true&stars=true)](https://nodei.co/npm/crazyhorse/) In your project's root directory use: ``` npm install crazyhorse ``` Usage: ====== Crazy horse will be one function call. It will take an array of sentence strings or a single string of sentences (I delimit/separate them with '. ' within the string). ```javascript var crazyhorse = require('crazyhorse') crazyhorse(list_of_sentences, function(remixed_sentence){ console.log(remixed_sentence) }) ``` We wil produce simple sentences at first of form [Noun Phrase][Verb Phrase]. Ideally eventually we will have a sentence structure and grammar like so: Sentence structure produced by crazyhorse: ``` [Interjection]||[CurseWord],[Sentence][Random Final Punctuation] [Noun Phrase] [Verb Phrase] [Verb][NounPhrase] [NounPhrase][Sentence] [Noun Phrase][Verb Phrase] ``` Simple Grammar: ``` [Noun Phrase] : [Article][Adjective || Adverb][Noun] || [Possesive][Noun || ProperNoun] [Verb Phrase] : [Verb][NounPhrase] [Sentence] : [Preposition][Sentence] || [Noun Phrase][Verb Phrase] ``` Test sentence list to be used as a list of sentences: ```javascript var list_of_sentences = "Water polo, or Water ball, is a team water sport. The playing team consists of six field players and one goalkeeper. The winner of the game is the team that scores the most goals. Game play involves swimming, treading water (using a sort of kicking motion known as 'eggbeater kick'), players passing the ball while being defended by opponents, and scoring by throwing the ball into a net defended by a goalie. 'Man-up' (or 'power play') situations occur frequently. Water polo, therefore, has strong similarities to the land-based game of team handball.The White House says now is not the time for Congress to get involved on Iran. When you skydive you have a backup altimeter that will deploy the chute at 1000 feet. All things normal, you'll live but have a rough landing if you are unconscious. The dangerous parts are what obstacles are there. You don't jump directly over the landing zone. You fly to it under canopy with the winds. Without guidance, thing like that river become deadly for an unconscious skydiver. It is also all over in today's Championship game as goals from Bradley Johnson and Lewis Grabban secure Norwich a league double over local rivals Ipswich. College admission season ignites deep anxieties for Asian American families, who spend more than any other demographic on education. At elite universities across the U.S., Asian Americans form a larger share of the student body than they do of the population as a whole. And increasingly they have turned against affirmative action policies that could alter those ratios, and accuse admissions committees of discriminating against Asian American applicants. Last year, a rumor that Harvard University would stop accepting any more Asian American students from San Marino High School spread like a trending hashtag. Just 12 hours ago, I posted a brief piece about the continuing Europtechnopanic in Germany and the effort of publishers there to blame their every trouble on Google—even the so-called sin of free content and the price of metaphoric wurst. Tomorrowland: In Silicon Valley, a new elite doesn’t just want to determine what we consume but how we live. They want to change the world and accept no regulation. Must we stop them? It is nothing less than prewar propaganda, trying to stir up a populace against a boogeyman enemy in hopes of goading politicians to action to stop these people. If anyone would know better, you’d think they would. Schade." ``` Ramblings: So one function well need to have is the ability to remember where words came from after they've been mixed. In otherwords save an array with references to the original tweet. The selection of words process should not be totally random. We should allow users to place weights on different tags returned by pos node module. This also lends itself well to machine learning in the future. Basically the wrong part of speech will never replace the wrong one( even if the right one has a 0 weight) but in a sentence where either could be used the part tag with higher weight will supercede. POS tags: ``` CC Coord Conjuncn and,but,or CD Cardinal number one,two DT Determiner the,some EX Existential there there FW Foreign Word mon dieu IN Preposition of,in,by JJ Adjective big JJR Adj., comparative bigger JJS Adj., superlative biggest LS List item marker 1,One MD Modal can,should NN Noun, sing. or mass dog NNP Proper noun, sing. Edinburgh NNPS Proper noun, plural Smiths NNS Noun, plural dogs POS Possessive ending 's PDT Predeterminer all, both PP$ Possessive pronoun my,one's PRP Personal pronoun I,you,she RB Adverb quickly RBR Adverb, comparative faster RBS Adverb, superlative fastest RP Particle up,off SYM Symbol +,%,& TO 'to' to UH Interjection oh, oops VB verb, base form eat VBD verb, past tense ate VBG verb, gerund eating VBN verb, past part eaten VBP Verb, present eat VBZ Verb, present eats WDT Wh-determiner which,that WP Wh pronoun who,what WP$ Possessive-Wh whose WRB Wh-adverb how,where , Comma , . Sent-final punct . ! ? : Mid-sent punct. : ; " $ Dollar sign $ # Pound sign # " quote " ( Left paren ( ) Right paren ) ``` Structure of the Buckets: ```JSON { "CC":["and", "or", "but"], . . . "DT":["the", "some"], "NN":["dog", "cat", "frog"], "NNP":["Edinburgh", "Jack"], . . . } ``` Nitpicked the most impactful curse words from: https://github.com/reimertz/curse-words