comments = raw_comments.split("</comment>")
courses += [[title, comments]]
return courses
"""
nltk.download('punkt')
nltk.download('stopwords')
nltk.download('maxent_treebank_pos_tagger')
"""
courses = parse_course_file("2014QComments")
positive = ["doable"]
negative = ["difficult", "hard", "work"]
analyzer = Analyze.SentimentAnalysis(positive, negative)
for course_num, course in enumerate(courses):
# Nouns and adjectives, run nltk.help.upenn_tagset() to see all possible tags
# pos = ["JJ", "JJR", "JJS", "NN", "NNP", "NNPS", "NNS"]
pos = ["NN", "NNP", "NNPS", "NNS"]
window = 2
sentences = []
custom_stop = [
"course", "class", "this", "will", "in", "you", "make", "sure",
"expect"
]
min_keyword_len = 4
key_phrases = key_phrases_for_course(course, pos, window, custom_stop,
min_keyword_len)
