dsad

import nltk
from nltk.classify.scikitlearn import SklearnClassifier
from sklearn.svm import LinearSVC
from sklearn.naive_bayes import MultinomialNB, BernoulliNB
from sklearn.neighbors import KNeighborsClassifier
from nltk.classify import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier
import matplotlib.pyplot as plt
from nltk.corpus import movie_reviews
import random

# създаване на списък от всички документи и тяхната категория
docs = [(list(movie_reviews.words(fileid)), category)
        for category in movie_reviews.categories()
        for fileid in movie_reviews.fileids(category)]

random.seed(43)
random.shuffle(docs)

# всички думи
all_words = []
for w in movie_reviews.words():
    all_words.append(w.lower())

all_words = nltk.FreqDist(all_words)

# топ 3000 думи
word_features = list(all_words.keys())[:3000]

def find_features(document):
    words = set(document)
    features = {}
    for w in word_features:
        features[w] = (w in words)
    return features

featuresets = [(find_features(rev), category) for (rev, category) in docs]

training_set = featuresets[:1900]
testing_set = featuresets[1900:]

# Бернулиев NB
BNB_classifier = SklearnClassifier(BernoulliNB())
BNB_classifier.train(training_set)
BNB_accuracy = nltk.classify.accuracy(BNB_classifier, testing_set) * 100
print("BernoulliNB accuracy percent:", BNB_accuracy)

# =========================
# МУЛТИНОМИАЛЕН NB (49-52)
MNB_classifier = SklearnClassifier(MultinomialNB())
MNB_classifier.train(training_set)
MNB_accuracy = nltk.classify.accuracy(MNB_classifier, testing_set) * 100
print("MultinomialNB accuracy percent:", MNB_accuracy)

# =========================
# SVM (54-57)
SVM_classifier = SklearnClassifier(LinearSVC())
SVM_classifier.train(training_set)
SVM_accuracy = nltk.classify.accuracy(SVM_classifier, testing_set) * 100
print("LinearSVC accuracy percent:", SVM_accuracy)

# =========================
# KNN (59-62)
KNN_classifier = SklearnClassifier(KNeighborsClassifier())
KNN_classifier.train(training_set)
KNN_accuracy = nltk.classify.accuracy(KNN_classifier, testing_set) * 100
print("KNN accuracy percent:", KNN_accuracy)

# =========================
# RANDOM FOREST (64-67)
RF_classifier = SklearnClassifier(RandomForestClassifier())
RF_classifier.train(training_set)
RF_accuracy = nltk.classify.accuracy(RF_classifier, testing_set) * 100
print("Random Forest accuracy percent:", RF_accuracy)

# =========================
# ADA BOOST (69-72)
AB_classifier = SklearnClassifier(AdaBoostClassifier())
AB_classifier.train(training_set)
AB_accuracy = nltk.classify.accuracy(AB_classifier, testing_set) * 100
print("AdaBoost accuracy percent:", AB_accuracy)

# =========================
# ДИАГРАМА (74-82)
classifiers = ['BNB', 'MNB', 'LSVC', 'KNN', 'RF', 'AB']
accuracies = [BNB_accuracy, MNB_accuracy, SVM_accuracy, KNN_accuracy, RF_accuracy, AB_accuracy]

plt.bar(classifiers, accuracies)
plt.xlabel("Classifiers")
plt.ylabel("Accuracy")
plt.title("Accuracy of different classifiers")
plt.show()