불균형 데이터셋

인구분포는 0~100세까지 어느정도는 고르게 퍼져있는 상태라고 치자. 그런데, 어떤 앱의 데이터를 분석하고자 했는데 특정한 앱의 연령대가 20대라고 하자. 그리고 이 앱의 데이터로 전체 인구의 특성 (모수)를 파악하고자 해보자. 일단 불균형한 샘플때문에 거의 인구가 20대로 수렴하는 현상이 생긴다.

랜덤하게 나눈 훈련셋과 테스트셋을 대상으로는 불균형 데이터셋을 그렇게 수정하지 않아도 어느정도 정확도가 나오는편이다. (보통 1.0~1.5배 뻥튀기 사이에서는 정확도가 늘어나는 경우가 종종있기도 하다.)

그런데, 우리가 완전히 보지않은 데이터셋은 이와 다른 분포를 지니고 있다. 그렇다고 한다고 해도 훈련셋 자체가 불균형 데이터셋으로 만들어져있어, 현실데이터에서도 한쪽으로 치우치는 결과가 나올 수 있다. 결국 데이터가 아무리 많더라도 진짜 현실에서 맞춰야 하는 태스크에 맞춰서 데이터를 잘라내는게 오히려 도움이 될 수 있다는 뜻.

16 thoughts on “불균형 데이터셋

  1. If some one wishes expert view concerning running a blog then i suggest
    him/her to pay a quick visit this weblog, Keep up
    the fastidious work.

    Here is my homepage them g

  2. My spouse and I stumbled over here coming from
    a different website and thought I should check things out.
    I like what I see so now i’m following you.
    Look forward to exploring your web page again.

    Feel free to surf to my web site; g if

  3. Have you ever thought about including a
    little bit more than just your articles? I mean, what you say
    is fundamental and all. However think of if you added some great photos or video clips to give your posts more, “pop”!
    Your content is excellent but with pics and clips, this website could certainly be one of the best in its field.
    Terrific blog!

    Have a look at my web blog: g for

  4. Fantastic goods from you, man. I have take
    note your stuff previous to and you are simply extremely fantastic.
    I actually like what you have received right here, really like what you’re stating and the way wherein you
    are saying it. You make it enjoyable and you still care for to
    keep it sensible. I can’t wait to learn far
    more from you. That is actually a terrific web site.

    Feel free to surf to my web page :: with g

  5. My partner and I stumbled over here from a different
    web page and thought I may as well check things out. I like what I
    see so now i’m following you. Look forward to looking
    at your web page for a second time.

    Also visit my webpage g rsacwgxy

  6. Hi! I’ve been reading your blog for some time now and finally got the courage to go ahead and give you
    a shout out from Houston Tx! Just wanted to tell you keep
    up the excellent work!

    Feel free to visit my web page; cbd oil (tinyurl.com)

  7. Actually when someone doesn’t be aware of then its up to other viewers that they will assist, so here it takes
    place.

    my web blog – cbd oil (detsad10.by)

  8. My partner and I absolutely love your blog and find
    the majority of your post’s to be exactly I’m looking for.
    can you offer guest writers to write content to suit
    your needs? I wouldn’t mind publishing a post or elaborating on many of
    the subjects you write related to here. Again, awesome blog!

    Take a look at my blog cbd oil that works 2020

  9. A fascinating discussion is definitely worth comment.
    I do think cbd oil that works 2020 you should write more on this subject, it may not be a taboo matter but typically people do not discuss these issues.
    To the next! Many thanks!!

  10. I’m truly enjoying the design and layout of your blog.

    It’s a very easy on the eyes which makes it much more enjoyable for me
    to come here and visit more often. Did you hire out
    a designer to create your theme? Exceptional work!

    My page :: cbd oil that works 2020

  11. buy cialis online new zealand [url=https://bbuycialisss.com/]when will cialis be generic[/url] Viagra Nuchternen Magen Buy Cialis Amoxicillin For Sore Throat

댓글 남기기