Adding additional information in content-based recommendations

I have a book dataset where 100 users have rated the books as like/dislike. Each observation with features
Table1 : [‘user_id’,’book_name’, ‘book_genre’,’author’,’date_published’,’like/dislike’]

These users were asked to list down their favorite authors and genres, they are stored in a different table as
Table2: [‘user_id’, ‘authors_liked’, ‘authors_disliked’, ‘genre_liked’, ‘genre_disliked’].
where the fields are lists of authors and genres liked/disliked.

Then there is the books database with information on all the available books(100,000 entries)
Table3: [‘book_id’,’book_name’, ‘book_genre’,’author’,’date_published’]

Question: Given a ‘book_name’ and ‘user_id’, predict if the user likes or dislikes it.

Currently, I have trained a classifier for Table1.
I have applied cosine similarity in Table2 and Table3.
I’m predicting the book as ‘likes’ if the output of the classifier is ‘like’ and the similarity is >0.5

Is there a better/standard way to incorporate Table2 information into the recommendation system?