在多对多连接表中，如何计算两个“所有者”共享的条目数？

Question

I have a list of movies and a list of tropes. To calculate the similarity between two movies, I am using cosine differences . If all the weights are even, then it simplifies pretty well:

similarity =

(number of shared tropes between both movies)
/
(SQRT(number of tropes from movie 1) + SQRT(number of tropes from movie 2))

For example, if movie 1 has tropes 1, 3, and 4, and movie 2 has tropes 1, 4, 6, and 7, then there would be two tropes shared between them and the similarity would be

2 / (SQRT(3) + SQRT(4)) = 2 / 3.73... = 0.54

My MySQL tables are pretty standard:

movies:
- id
- ...

tropes:
- id
- ...

movie_tropes:
- movie_id
- trope_id

I can easily count the number of tropes for a movie:

SELECT count(distinct trope_id) from movie_tropes where movie_id = 1;
SELECT count(distinct trope_id) from movie_tropes where movie_id = 2;

I am a little out of practice with SQL. Is there a simple join-y way to count the number of trope_ids that occur for both movie 1 and movie 2 in this join table?

Answer 1

Is there a simple way to count the number of trope_ids that occur for both movie 1 and movie 2?

You can self-join:

select count(distinct trope_id)
from movie_tropes t1
inner join movie_tropes t2 on t2.trope_id = t1.trope_id
where t1.movie_id = 1 and t2.movie_id = 2

But overall, you can compute the three base figures at once with two levels of aggregation. I would recommend:

select 
    sum(has_1) as cnt_1,            -- count of distinct tropes for movie 1
    sum(has_2) as cnt_2,            -- count of distinct tropes for movie 2
    sum(has_1 and has_2) as cnt_both  -- count of distinct tropes for both movies
from (
    select max(movie_id = 1) has_1, max(movie_id = 2) as has_2
    from movie_tropes t
    where movie_id in (1, 2)
    group by trope_id
) t