NLTK庫并不是一個主要用于機器學習模型選擇的工具，它更多用于自然語言處理任務。但是，可以結合NLTK庫和其他機器學習庫（如scikit-learn）來進行模型選擇。以下是一個使用NLTK和scikit-learn庫進行模型選擇的示例：

1. 導入必要的庫：

import nltk
from nltk.classify.scikitlearn import SklearnClassifier
from sklearn.naive_bayes import MultinomialNB
from sklearn.svm import SVC
from sklearn.model_selection import cross_val_score

2. 加載數據集，并進行特征提取和數據準備：

from nltk.corpus import movie_reviews

documents = [(list(movie_reviews.words(fileid)), category)
             for category in movie_reviews.categories()
             for fileid in movie_reviews.fileids(category)]

# Shuffle the documents
import random
random.shuffle(documents)

all_words = nltk.FreqDist(w.lower() for w in movie_reviews.words())
word_features = list(all_words)[:2000]

def document_features(document):
    document_words = set(document)
    features = {}
    for word in word_features:
        features['contains({})'.format(word)] = (word in document_words)
    return features

featuresets = [(document_features(d), c) for (d,c) in documents]

3. 劃分數據集為訓練集和測試集，并使用交叉驗證評估不同模型的性能：

train_set, test_set = featuresets[100:], featuresets[:100]

nb_classifier = SklearnClassifier(MultinomialNB())
svm_classifier = SklearnClassifier(SVC())

nb_scores = cross_val_score(nb_classifier, train_set, cv=5)
svm_scores = cross_val_score(svm_classifier, train_set, cv=5)

print("Naive Bayes Classifier Accuracy:", nb_scores.mean())
print("SVM Classifier Accuracy:", svm_scores.mean())

通過以上步驟，可以使用NLTK和scikit-learn庫進行模型選擇，并選擇性能最佳的模型進行進一步優化和預測。