nlpsum/nlp-node-master/inflection/singularize.js

Powerful text summarization algorithms from research papers and dedicated research.

var singularize = (function() {
  //Originally by david huynh 2010
  //http://www.freebase.com/appeditor/#!path=//cubed.dfhuynh.user.dev/index
  //Algorithm is adopted from
  //http://www.csse.monash.edu.au/~damian/papers/HTML/Plurals.html
  //Adapted by spencer kelly @spencermountain
  var singularize = function(text) {

    if (text.match(' ')) { //multiple words
      var words = text.split(' ');
      var last = words[words.length - 1];
      var firsts = words.slice(0, -1);
      return firsts.join(" ") + ' ' + singularize(last);
    }

    var prepositions = {
      "about": 1,
      "above": 1,
      "across": 1,
      "after": 1,
      "against": 1,
      "along": 1,
      "among": 1,
      "around": 1,
      "at": 1,
      "before": 1,
      "behind": 1,
      "below": 1,
      "beneath": 1,
      "beside": 1,
      "between": 1,
      "beyond": 1,
      "but": 1,
      "by": 1,
      "despite": 1,
      "down": 1,
      "during": 1,
      "except": 1,
      "for": 1,
      "from": 1,
      "in": 1,
      "inside": 1,
      "into": 1,
      "like": 1,
      "near": 1,
      "of": 1,
      "off": 1,
      "on": 1,
      "onto": 1,
      "out": 1,
      "outside": 1,
      "over": 1,
      "past": 1,
      "since": 1,
      "through": 1,
      "throughout": 1,
      "till": 1,
      "to": 1,
      "toward": 1,
      "under": 1,
      "underneath": 1,
      "until": 1,
      "up": 1,
      "upon": 1,
      "with": 1,
      "within": 1,
      "without": 1
    };



    var userDefinedNouns = [{
        "p": "people",
        "s": "person"
      }, {
        "p": "tornadoes",
        "s": "tornado"
      }, {
        "p": "churches",
        "s": "church"
      }, {
        "p": "countries",
        "s": "country"
      }, {
        "p": "cities",
        "s": "city"
      }, {
        "p": "companies",
        "s": "company"
      }, {
        "p": "monkies",
        "s": "monkey"
      }, {
        "p": "donkies",
        "s": "donkey"
      }, {
        "p": "mysteries",
        "s": "mystery"
      }, {
        "p": "authors",
        "s": "author"
      }
    ];

    // Table A.1
    var irregularNouns = {
      "beef": {
        anglicized: "beefs",
        classical: "beeves"
      },
      "brother": {
        anglicized: "brothers",
        classical: "brethren"
      },
      "child": {
        anglicized: null,
        classical: "children"
      },
      "cow": {
        anglicized: null,
        classical: "kine"
      },
      "ephemeris": {
        anglicized: null,
        classical: "ephemerides"
      },
      "genie": {
        anglicized: null,
        classical: "genii"
      },
      "money": {
        anglicized: "moneys",
        classical: "monies"
      },
      "mongoose": {
        anglicized: "mongooses",
        classical: null
      },
      "mythos": {
        anglicized: null,
        classical: "mythoi"
      },
      "octopus": {
        anglicized: "octopuses",
        classical: "octopodes"
      },
      "ox": {
        anglicized: null,
        classical: "oxen"
      },
      "soliloquy": {
        anglicized: "soliloquies",
        classical: null
      },
      "trilby": {
        anglicized: "trilbys",
        classical: null
      }
    };

    var uninflectedSuffixes = [
        "fish", "ois", "sheep", "deer", "pox", "itis"
    ];

    // Table A.2
    var uninflectedNouns = {
      "bison": 1,
      "flounder": 1,
      "pliers": 1,
      "bream": 1,
      "gallows": 1,
      "proceedings": 1,
      "breeches": 1,
      "graffiti": 1,
      "rabies": 1,
      "britches": 1,
      "headquarters": 1,
      "salmon": 1,
      "carp": 1,
      "herpes": 1,
      "scissors": 1,
      "chassis": 1,
      "high-jinks": 1,
      "sea-bass": 1,
      "seabass": 1,
      "clippers": 1,
      "homework": 1,
      "series": 1,
      "cod": 1,
      "innings": 1,
      "shears": 1,
      "contretemps": 1,
      "jackanapes": 1,
      "species": 1,
      "corps": 1,
      "mackerel": 1,
      "swine": 1,
      "debris": 1,
      "measles": 1,
      "trout": 1,
      "diabetes": 1,
      "mews": 1,
      "tuna": 1,
      "djinn": 1,
      "mumps": 1,
      "whiting": 1,
      "eland": 1,
      "news": 1,
      "wildebeest": 1,
      "elk": 1,
      "pincers": 1,

      "moose": 1,
      "shrimp": 1,
      "hoi polloi": 1,
      "riffraff": 1,
      "rabble": 1
    };
    var inflectionCategories = [{ // Table A.10
        from: "a",
        to: "ae",
        words: ["alumna", "alga", "vertebra"]
      }, {
        // Table A.11
        from: "a",
        anglicized: "as",
        classical: "ae",
        words: ["abscissa", "amoeba", "antenna", "aurora", "formula", "hydra", "hyperbola", "lacuna", "medusa", "nebula", "nova", "parabola"]
      }, {
        // Table A.12
        from: "a",
        anglicized: "as",
        classical: "ata",
        words: ["anathema", "bema", "carcinoma", "charisma", "diploma", "dogma", "drama", "edema", "enema", "enigma", "gumma", "lemma", "lymphoma", "magma", "melisma", "miasma", "oedema", "sarcoma", "schema", "soma", "stigma", "stoma", "trauma"]
      }, {
        // Table A.13
        from: "en",
        anglicized: "ens",
        classical: "ina",
        words: ["stamen", "foramen", "lumen"]
      }, {
        // Table A.14
        from: "ex",
        to: "ices",
        words: ["codex", "murex", "silex"]
      }, {
        // Table A.15
        from: "ex",
        anglicized: "exes",
        classical: "ices",
        words: ["apex", "cortex", "index", "latex", "pontifex", "simplex", "vertex", "vortex"]
      }, {
        // Table A.16
        from: "is",
        anglicized: "ises",
        classical: "ides",
        words: ["iris", "clitoris"]
      }, {
        // Table A.17
        from: "o",
        to: "os",
        words: ["albino", "archipelago", "armadillo", "commando", "ditto", "dynamo", "embryo", "fiasco",
            "generalissimo", "ghetto", "guano", "inferno", "jumbo", "lingo", "lumbago", "magneto",
            "manifesto", "medico", "octavo", "photo", "pro", "quarto", "rhino", "stylo"
        ]
      }, {
        // Table A.18
        from: "o",
        anglicized: "os",
        classical: "i",
        words: ["alto", "basso", "canto", "contralto", "crescendo", "solo", "soprano", "tempo"]
      }, {
        // Table A.19
        from: "on",
        to: "a",
        words: ["aphelion", "asyndeton", "criterion", "hyperbaton", "noumenon", "organon", "perihelion", "phenomenon", "prolegomenon"]
      }, {
        // Table A.20
        from: "um",
        to: "a",
        words: ["agendum", "bacterium", "candelabrum", "datum", "desideratum", "erratum", "extremum", "stratum", "ovum"]
      }, {
        // Table A.21
        from: "um",
        anglicized: "ums",
        classical: "a",
        words: ["aquarium", "compendium", "consortium", "cranium", "curriculum", "dictum", "emporium", "enconium", "gymnasium", "honorarium",
            "interregnum", "lustrum", "maximum", "medium", "memorandum", "millenium", "minimum", "momentum", "optimum", "phylum",
            "quantum", "rostrum", "spectrum", "speculum", "stadium", "trapezium", "ultimatum", "vacuum", "velum"
        ]
      }, {
        // Table A.22
        from: "us",
        anglicized: "uses",
        classical: "i",
        words: ["focus", "fungus", "genius", "incubus", "nimbus", "nucleolus", "radius", "stylus", "succubus", "torus", "umbilicus", "uterus"]
      }, {
        // Table A.23
        from: "us",
        anglicized: "uses",
        classical: "us",
        words: ["apparatus", "cantus", "coitus", "hiatus", "impetus", "nexus", "plexus", "prospectus", "sinus", "status"]
      }, {
        // Table A.24
        from: "",
        to: "i",
        words: ["afreet", "afrit", "efreet"]
      }, {
        // Table A.25
        from: "",
        to: "im",
        words: ["cherub", "goy", "geraph"]
      }
    ];

    function suffix(text, s) {
      return text.length >= s.length && text.substring(text.length - s.length) == s;
    }

    function capIfCap(s, s2) {
      if (typeof s == "string") {
        var isCap = s2.charAt(0).toLowerCase() != s2.charAt(0);
        return isCap ? (s.charAt(0).toUpperCase() + s.substr(1)) : s;
      } else {
        var a = [];
        for (var i in s) {
          var s3 = s[i];
          a.push(capIfCap(s3, s2));
        }
        return a;
      }
    }


    function inflection(text, from, to) {
      return text.substring(0, text.length - from.length) + to;
    }


    function isOneOf(c, chars) {
      return chars.indexOf(c) >= 0;
    }

    function isVowel(c) {
      return isOneOf(c, "aeiou");
    }



    var text2 = text.toLowerCase();



    for (var o in userDefinedNouns) {
      if (userDefinedNouns[o].p == text) {
        return userDefinedNouns[o].s;
      }
    }

    for (var singular in irregularNouns) {
      var entry = irregularNouns[singular];
      if (entry.anglicized === text2 || entry.classical === text2) {
        return capIfCap(singular, text);
      }
    }

    for (var s in uninflectedSuffixes) {
      if (suffix(text2, s)) {
        return text;
      }
    }

    if (uninflectedNouns && uninflectedNouns[text2]) {
      return text;
    }

    var checkWords = function(from, to, words) {
      if (suffix(text, to)) {
        var prefix = text.substring(text.length - to.length);
        var text3 = prefix + entry.from;
        for (var word in words) {
          if (text3 === word) {
            return capIfCap(text3, text);
          }
        }
      }
      return null;
    };

    for (var e in inflectionCategories) {
      var entry = inflectionCategories[e];
      var text3 =
        ("to" in entry && checkWords(entry.from, entry.to, entry.words)) ||
        ("anglicized" in entry && checkWords(entry.from, entry.anglicized, entry.words)) ||
        ("classical" in entry && checkWords(entry.from, entry.classical, entry.words));

      if (text3 != null && typeof text3 == "string") {
        return text3;
      }
    }

    for (var prep in prepositions) {
      var n = text.indexOf(" " + prep + " ");
      if (n > 0) {
        var prefix = text.substring(0, n);
        var r = singularize(prefix);
        if (r != null) {
          return r + " " + prep + " " + text.substr(n + prep.length + 2);
        } else {
          return null;
        }
      }
      n = text.indexOf("-" + prep + "-");
      if (n > 0) {
        var prefix = text.substring(0, n);
        var r = singularize(prefix);
        if (r != null) {
          return r + "-" + prep + "-" + text.substr(n + prep.length + 2);
        } else {
          return null;
        }
      }
    }

    var j = text.lastIndexOf(" ");
    if (j > 0) {
      var r = singularize(text.substring(j + 1));
      if (r != null) {
        return text.substring(0, j + 1) + r;
      } else {
        return null;
      }
    }

    if (suffix(text, "xes") || suffix(text, "ses")) {
      return text.substring(0, text.length - 2);
    }
    if (suffix(text, "s") && !suffix(text, "ss")) {
      return text.substring(0, text.length - 1);
    }


    return text;
  }


  //console.log(exports.singularize("george soros"));
  //console.log(exports.singularize("mama cass"));

  //var start = new Date().getTime();
  //console.log(exports.singularize('earthquakes'));
  //console.log(new Date().getTime() - start);



  // export for AMD / RequireJS
  if (typeof define !== 'undefined' && define.amd) {
    define([], function() {
      return singularize;
    });
  }
  // export for Node.js
  else if (typeof module !== 'undefined' && module.exports) {
    module.exports = singularize;
  }

  return singularize;


})()