MovieRecommendation/code.py at master · prithuls/MovieRecommendation · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
from typing import Union
import pandas as pd
import difflib
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics.pairwise import cosine_similarity

movie = pd.read_csv('moviedata.csv')

features = ['keywords', 'cast', 'genres', 'director', 'vote_average']

for feature in features:
    movie[feature] = movie[feature].fillna('')


def combine_feartures(row):
    try:
        result = row['keywords'] + ' ' + row['cast'] + ' ' + row['genres'] + ' ' + row['director'] + ' ' + str(
            row['vote_average'])
        return result
    except:
        print("Error")


movie["combined_features"] = movie.apply(combine_feartures, axis=1)


def title_from_index(index):
    return movie[movie.index == index]["title"].values[0]


def index_from_title(title):
    title_list = movie["title"].tolist()
    common = difflib.get_close_matches(title, title_list, 1)
    title_sim: Union[str, bytes] = common[0].title()
    print(title_sim)
    return movie[movie.title == title_sim]["index"].values[0]


cv = CountVectorizer()
count_matrix = cv.fit_transform(movie["combined_features"])
cosine_sim = cosine_similarity(count_matrix)

user_movie = input("Enter Movie Name:\t")
movie_index = index_from_title(user_movie)

similar_movies = list(enumerate(cosine_sim[movie_index]))
similar_movies_sorted = sorted(similar_movies, key=lambda x: x[1], reverse=True)

i = 0
print("Other movies suggestion:")
for rec_movie in similar_movies_sorted:
    if i != 0:
        print(i, title_from_index(rec_movie[0]), sep=". ")
    i += 1
    if i > 50:
        break