유사도 찾기

Python으로 추천 엔진 만들기

Rob O'Callaghan

Director of Data

아이템 기반 추천

사용자 기반에서 아이템 기반으로

print(user_ratings_pivot):

          The Great Gatsby    The Catcher in the Rye    Fifty Shades of Grey                    
User_233               0.0                       0.0                     0.0
User_651               0.0                       0.5                    -0.5
User_965               0.5                      -0.5                     0.0
     ...               ...                       ...                     ...

book_ratings_pivot = user_ratings_pivot.T
print(book_ratings_pivot)

                  User_233                  User_651                User_965                    
The Great Gatsby       0.0                       0.0                     0.5
The Catcher in the Rye 0.0                       0.5                    -0.5
Fifty Shades of Grey   0.0                      -0.5                     0.0
                 ...   ...                       ...                     ...

코사인 유사도

book_ratings_pivot:

                  User_233                  User_651                User_965                    
The Great Gatsby       0.0                       0.0                     0.5
The Catcher in the Rye 0.0                       0.5                    -0.5
Fifty Shades of Grey   0.0                      -0.5                     0.0
                 ...   ...                       ...                     ...

코사인 유사도

cosine_similarity(                                                                    ,
                                                                               )

코사인 유사도

cosine_similarity(book_ratings_pivot.loc['Lord of the Rings', :]                      ,
                  book_ratings_pivot.loc['The Hobbit', :]                      )

코사인 유사도

cosine_similarity(book_ratings_pivot.loc['Lord of the Rings', :].values               ,
                  book_ratings_pivot.loc['The Hobbit', :].values               )

코사인 유사도

cosine_similarity(book_ratings_pivot.loc['Lord of the Rings', :].values.reshape(1, -1),
                  book_ratings_pivot.loc['The Hobbit', :].values.reshape(1, -1))

0.43

cosine_similarity(book_ratngs.loc['Lord of the Rings', :].values.reshape(1, -1),
                  book_ratngs.loc['Twilight', :].values.reshape(1, -1))

-0.64

코사인 유사도

similarities = cosine_similarity(book_ratings_pivot)

cosine_similarity_df = pd.DataFrame(book_ratings_pivot,
                                    index=book_ratings_pivot.index,
                                    columns=book_ratings_pivot.index)

cosine_similarity_df.head()

          The Great Gatsby    The Catcher in the Rye    Fifty Shades of Grey                    
The Great Gatsby       1.0                       0.0                    -0.3
The Catcher in the Rye 0.0                       1.0                    -0.5
Fifty Shades of Grey  -0.3                      -0.5                     1.0
                 ...   ...                       ...                     ...

코사인 유사도

cosine_similarity_series = cosine_similarity_df.loc['The Hobbit']

ordered_similarities = cosine_similarity_series.sort_values(ascending=False)

print(ordered_similarities)

The Hobbit         1.00
Lord of the Rings  0.43
The Silmarillion   0.37
...

연습해 봅시다!

Python으로 추천 엔진 만들기