def prepare():
print("[prepareDB] Starting pepare database...")
startTime = time.time()
mongo = Mongo("movieRecommend")
# Store MovieLens data into database.
# runtime: (1~2hours)
movieLensParser.parse(mongo)
# Add ANEW all list into database.
# runtime: (0.05s)
anewParser.parse(mongo)
# Pre-computation
# runtime: (few seconds)
prepare_genres(mongo)
prepare_actors(mongo)
# all kinds of ranking
# runtime: (few seconds)
prepare_rankings(mongo)
# recommendations for all movies
# runtime: (1~2hours)
prepare_recommend(mongo)
print("[prepareDB] Done (%0.2fs)." % (time.time() - startTime))
def main():
mongo = Mongo("movieRecommend")
db_imdb = mongo.client["imdb"]
db_imdb["movies"].create_index([("title", pymongo.ASCENDING)])
print("[movieLensFullTitle] Created index for title in movies")
retrieve(mongo)
db_movieRecommend = mongo.client["movieRecommend"]
db_movieRecommend["movie"].create_index([("title_full", pymongo.ASCENDING)
])
print("[movieLensFullTitle] Created index for title_full in movie")
def __init__(self, user_name, exist=False):
self.name = user_name
self.mongo = Mongo("movieRecommend")
self.recommend = MovieRecommend(mongo)
if not exist:
# tag ids
self.tags = set()
# movie ids
self.movies = set()
else:
self.load_user()
def main():
textAnalytics = TextAnalytics(Mongo("movieRecommend"),
anew=True,
aylien=True)
# sentence = "Congrats to @HCP_Nevada on their health care headliner win"
# sentence = "b'I love you @iHeartRadio! I love you hooligans! love you Sriracha. I love you @LeoDiCaprio. Thinking of u @holyfield https://t.co/iPoHf03G4R'"
sentence = "The secret life of Walter Mitty is a fantastic movie"
print("[TweetAnalytics] Evaluating sentence: " + sentence)
score = textAnalytics.gain_sentiment(sentence)
print("[TweetAnalytics] Sentiment score: " + str(score))
hashtag = "Askthedragon"
# hashtag = "NothingTo1990sDoA-story"
# hashtag = "9/11"
# hashtag = "1980sWhereAreYou"
print(textAnalytics.get_words_from_hashtag(hashtag))
tweet = "DAMN! Glo-Zell To da No! Why I gotta be 4-2? HAHAHA! http://www.youtube.com/watch?v=aQoDEZI4ces Watch Glozell Snap on me AGAIN! #Damn"
# tweet = "@Ellichter if you make a left on Boo Boo lane youll end up @ the Ca Ca Mart. Thats where they sell a wide variety of Doo Doo Spread.. #ill"
# tweet = "This year for me is all about Touring and playing shows and i believe im going Everywhere! I cant wait to see you guys live & in concert!"
# tweet = "@TheEllenShow Thank you!! cant wait to see you in January.."
# tweet = "Performing in 5....4.....3.......2........."
print(textAnalytics.get_words_from_tweet(tweet))
def main():
mongo = Mongo()
db_imdb = mongo.client["imdb"]
db_imdb["movies"].create_index([("keywords", pymongo.ASCENDING)])
print("[keywordsCombine] Created index for keywords in movies")
collect_from_keywords(mongo.client) # 34 seconds
collect_from_tags(mongo.client) # 5 minutes
combine(mongo.client) # 8 seconds
db_integration = mongo.client["integration"]
db_integration["integrated_tag"].create_index([("tag", pymongo.ASCENDING)])
print("[keywordsCombine] Created index for tag in integrated_tag")
# not used due to inaccurate
# # fix_popularity(mongo.client) # 3 seconds
imdbPeopleIndex.build(mongo) # 3 minutes
store_people_name_only(mongo.client) # 36 seconds
reconstruct_tags(mongo.client) # 6 seconds
db_integration = mongo.client["integration"]
db_integration["normalized_tags"].create_index([("tag", pymongo.ASCENDING)
])
print("[keywordsCombine] Created index for tag in normalized_tags")
count_movies_with_tags(mongo.client) # 3 seconds
copy_movies(mongo.client) # 15 seconds
db_integration = mongo.client["integration"]
db_integration["copy_movies"].create_index([("imdbtitle",
pymongo.ASCENDING)])
print("[keywordsCombine] Created index for imdbtitle in copy_movies")
def main():
recommender = MovieRecommend(Mongo("movieRecommend"))
# # -----------------------------------------------------------------
# # unit test, input: User ID = 4
# print("[MovieRecommend] ***** Unit test for recommend_movies_for_user() *****")
# user_id = 4
# recommends = recommender.recommend_movies_for_user(user_id)
# recommender.print_recommend(recommends)
# # -----------------------------------------------------------------
# # unit test, input tags:
# # [28, 387, 599, 704, 794]
# # ["adventure", "feel-good", "life", "new york city", "police"]
# print("[MovieRecommend] ***** Unit test for recommend_movies_based_on_tags() *****")
# tags = [28, 387, 599, 704, 794]
# recommends = recommender.recommend_movies_based_on_tags(tags)
# recommender.print_recommend(recommends)
# print("[MovieRecommend] ***** Unit test for recommend_movies_based_on_tags() with tag contents input *****")
# tags = ["adventure", "feel-good", "life", "new york city", "police"]
# recommends = recommender.recommend_movies_based_on_tags(tags, tagid=False)
# recommender.print_recommend(recommends)
# # -----------------------------------------------------------------
# # unit test, input: Movie ID = 1 "Toy Story (1995)"
# print("[MovieRecommend] ***** Unit test for recommend_movies_for_movie() *****")
# movie_id = 1
# recommends = recommender.recommend_movies_for_movie(movie_id)
# recommender.print_recommend(recommends)
# # -----------------------------------------------------------------
# print("[MovieRecommend] ***** Unit test for recommend_movies_for_twitter() *****")
# user_screen_name = "BrunoMars"
# # user_screen_name = "LeoDiCaprio"
# # user_screen_name = "BarackObama"
# # user_screen_name = "sundarpichai"
# # user_screen_name = "BillGates"
# # user_screen_name = "jhsdfjak"
# recommends = recommender.recommend_movies_for_twitter(user_screen_name)
# # recommender.print_recommend(recommends)
# print(recommender.get_titles_by_mids(recommends))
# # -----------------------------------------------------------------
# print("[MovieRecommend] ***** Unit test for recommend_movies_for_twitter_integrated() *****")
# user_screen_name = "BrunoMars"
# # user_screen_name = "LeoDiCaprio"
# # user_screen_name = "BarackObama"
# # user_screen_name = "sundarpichai"
# # user_screen_name = "BillGates"
# # user_screen_name = "jhsdfjak"
# recommends = recommender.recommend_movies_for_twitter_integrated(user_screen_name)
# for recommend in recommends:
# print(recommend.encode("utf8"))
# # -----------------------------------------------------------------
# # unit test, input tags:
# # ["adventure", "feel good", "life", "new york city", "police"]
# print("[MovieRecommend] ***** Unit test for recommend_movies_based_on_tags_integrated() with tag contents input *****")
# tags = ["adventure", "feel good", "life", "new york city", "police"]
# recommends = recommender.recommend_movies_based_on_tags_integrated(tags)
# for recommend in recommends:
# print(recommend.encode("utf8"))
# # -----------------------------------------------------------------
# unit test for recommend_movies_based_on_history()
print(
"[MovieRecommend] ***** Unit test for recommend_movies_based_on_history() *****"
)
user_history = []
user_history.append("Toy Story (1995)")
user_history.append("Big Hero 6 (2014)")
user_history.append("X-Men: Days of Future Past (2014)")
user_history.append("The Lego Movie (2014)")
user_history.append("The Secret Life of Walter Mitty (2013)")
user_history.append("Death Note: Desu nto (2006)")
user_history.append("Zombieland (2009)")
user_history.append("Fifty Shades of Grey (2015)")
user_history.append("The Maze Runner (2014)")
recommends = recommender.recommend_movies_based_on_history(user_history)
recommends = recommender.get_titles_by_mids(recommends)
for recommend in recommends:
print(recommend.encode("utf8"))
def main():
mongo = Mongo("movieRecommend")
parse(mongo)
def main():
# Add ANEW all list into database.
# runtime: (0.05s)
mongo = Mongo("movieRecommend")
parse(mongo)
def main():
mongo = Mongo("movieRecommend")
retrieve(mongo)
def main():
mongo = Mongo("imdb")
build(mongo)
def main():
twitter = Tweepy(Mongo("movieRecommend"))
# twitter.get_rate_limit()
# twitter.extract_profile("LeoDiCaprio")
# twitter.extract_profile("BrunoMars")
twitter.extract_profile("jhsdfjak")