第1章: 準備運動

01.clj

;;; https://nlp100.github.io/ja/ch01.html

;; helper
(defn get-words [s]
  (->> (clojure.string/split s #"[\., ]")
       (filter #(not (empty? %)))))

(defn make-ngram [n coll]
  (if (> n (count coll))
    []
    (cons (take n coll) (make-ngram n (rest coll)))))

(def bi-gram (partial make-ngram 2))
(defn bi-gram-string [s]
  (->>
   (bi-gram (clojure.string/replace s #" " ""))
   (map #(apply str %))))

;; 01
(apply str (reverse "stressed"))

;; 02
(let [s "パタトクカシーー"]
  (apply str (take-nth 2 s)))

;; 03
(let [s "Now I need a drink, alcoholic of course, after the heavy lectures involving quantum mechanics."
      words (get-words s)]
  (map (fn [word] (count word)) words))

;; 04
(let [s "Hi He Lied Because Boron Could Not Oxidize Fluorine. New Nations Might Also Sign Peace Security Clause. Arthur King Can."
      one-char-indexes [1 5 6 7 8 9 15 16 19]
      words (get-words s)]
  (->>
   (map-indexed (fn [& args] args) words)
   (map (fn [[i w]]
          (let [take-n (if (some #(= (inc i) %) one-char-indexes) 1 2)]
            [(apply str (take take-n w)) (inc i)])))
   (into {})))

;; 05

;; word bi-gram
(let [s "I am an NLPer"
      words (get-words s)]
  (bi-gram words))

;; character bi-gram
(let [s "I am an NLPer"
      words (get-words s)]
  (->>
   (bi-gram (clojure.string/replace s #" " ""))
   (map #(apply str %))))

;; 06
(let [s1 "paraparaparadise"
      s2 "paragraph"
      x (into #{} (bi-gram-string s1))
      y (into #{} (bi-gram-string s2))]
  (let [union (clojure.set/union x y) ; 和集合
        intersection (clojure.set/intersection x y) ; 積集合
        difference (clojure.set/difference x y)] ; 差集合
    (clojure.pprint/pprint union)
    (clojure.pprint/pprint intersection)
    (clojure.pprint/pprint difference)
    (clojure.pprint/pprint(contains? x "se"))
    (clojure.pprint/pprint(contains? y "se"))))

;; 07

(defn hoge [x y z]
  (str x "時の" y "は" z))

(println (hoge 12 "気温" 22.4))

;; 08

(defn cipher [s]
  (->> (map (fn [c]
              (if (Character/isLowerCase c)
                (char (- 219 (int c)))
                c)) s)
       (apply str)))

(cipher (cipher "abcdefghijklmnopqlstuvwxyz"))

;; 09
(let [s "I couldn't believe that I could actually understand what I was reading : the phenomenal power of the human mind ."
      words (clojure.string/split s #" ")]
  (map (fn [word]
         (if (> (count word) 4)
           (let [first-char (nth word 0)
                 last-char (nth word (dec (count word)))
                 interval-char (apply str (shuffle (drop-last (drop 1 word))))]
             (str first-char
                  interval-char
                  last-char))
           word))
       words))