Why does Codechef drop ratings solely based on Moss scores?

april16
avoid_cheating
cheating
moss

#1

@admin,
It’s been a day, you’ve
removed my solutions without even informing me, you don’t accept calls
and I am yet to receive a reply to my emails. This is highly
unprofessional on your part. You need to consider urgent issues faster.

Ratings don’t really matter to me. But Codechef Long would be
wonderful without this “guilty until proved innocent” mechanism.

From https://theory.stanford.edu/~aiken/moss/

"Moss is not a system for completely automatically detecting
plagiarism. Plagiarism is a satement that someone copied code
deliberately without attribution, and while Moss automatically detects
program similarity, it has no way of knowing why codes are similar. It
is still up to a human to go and look at the parts of the code that
Moss highlights and make a decision about whether there is plagiarism
or not. One way of thinking about what Moss provides is that it saves
teachers and teaching staff a lot of time by pointing out the parts of
programs that are worth a more detailed examination. But once someone
has looked at those portions of the programs, it shouldn’t matter
whether the suspect code was first discovered by Moss or by a human;
the case that there was plagiarism should stand on its own.

In particular, it is a misuse of Moss to rely solely on the similarity
scores. These scores are useful for judging the relative amount of
matching between different pairs of programs and for more easily
seeing which pairs of programs stick out with unusual amounts of
matching. But the scores are certainly not a proof of plagiarism.
Someone must still look at the code."

  1. I have read your Code of Conduct and abide by it.
  2. I did not copy neilkakkar’s solution, though our solutions are
    surprisingly similar, probably because we have similar variable
    naming/indentation habits. Although this is highly improbable, it’s
    not impossible.
  3. I think, almost any other competitive programmer, who didn’t know a
    better solution would have made a similar submissions.
  4. Do you really think I’d copy a solution for 10 points? I spent
    sleepless nights trying to frame a 100-scoring solution to this
    problem.
  5. If you still think I copied the solution, go ahead and drop my
    rating. It would have been my first Long Challenge goody, had my
    solution not been disqualified. But, because I can’t really establish
    that my solution wasn’t copied and that guy had made his submission a
    day before me, GO AHEAD.

And please don’t delete this post, if you don’t want it all over the internet.

Regards,
Arpan


#2

Hey arpanb8, Firstly, let us apologize for the delay in responding to your email. We will check your email and respond at the earliest. Secondly, we never delete any posts (except the spam posts). You will be getting a reply from us soon.


#3

Happened with me too… Some innocent people’s accounts are affected ,and nothing happened to few cheaters.


#4

Hello arpanb8, I am facing the same problem. I sent a lengthy mail today in an attempt to assert that just because my code is similar to someone else’s, it does not mean that either of us has cheated. I received a reply 3 hours ago informing me that they have decided to “revisit the MOSS benchmarks and set up new ones”. Hopefully ratings will be restored and the plagiarism detection system improved for good. Let’s not give up on a good site like Codechef :slight_smile:


#5

I would like to add in my “grievances” too.

I have been wrongly penalized twice now. Once, in the October 15 long challenge, and then again during the February 16 long challenge. Both times, it has been false accusations of plagiarism. The first time it happened, I know how sad I was, because being quite fresh to competitive coding, I had solved 5 questions completely, and my rank was around 500, and that was a very happy achievement for me. Ratings do not matter to me at the end of the day, but disqualifying the solutions was like a hit to the effort I had put in those 10 days. I mailed the guys like 5 times (I have them in my mailbox still). None of them have been replied to. I made calls; no one picked. I tried messaging them on Facebook; fail. I even put it up here, where others like @ceilks accepted that it wasn’t my fault at all. The same very thing happens again in February 16, but this time I don’t try to email them, because I know it will be of no use.

The main thing was the solution that was penalized for was such a straight forward implementation, that the possibility of the two solutions matching was very high, though unfair. On top of that, the other person whose code was matched, is not even of my nation!. I see some of my friends getting penalized this month too. I feel very bad for them and for their efforts not being respected.


#6

Hello arpanb8,

Many plaigarism cases occurs on simple and cakewalk problems,
such solutions are of less than 15 lines ,
often people get caught.

So one thing they can do is do plaigarims check only on problems where number of submissions is less than 2k or having a low accuracy, that is where people tend to copy , not on simple problems.

And having said that moss doesnt check the flow of the program rather does string matching ,
many people would reorder their program , change variable name (Find and replace :stuck_out_tongue: ) and still get away with it.


#7

@arpanb8
I agree with your points about MOSS.
Codechef does not seem very fair to me when it comes to plagiarism cases. I’ve reported many of the plagiarism cases in the past through e-mail and posting it here :

https://discuss.codechef.com/questions/75155/plagiarism-in-codecrck-sept15-long-challenge

No action has ever been taken on these reports.

This time too, have a look at these two solutions :

https://www.codechef.com/viewsolution/9852757

https://www.codechef.com/viewsolution/9902596

The only difference I could figure out was different order of some statements and different names for variables. The first one was penalised and the second one fooled the MOSS.

PS : I’m sure codechef won’t even look at these two solutions, because they tend to believe that the software results are more accurate than the human observations.


#8

I’d like to add to your comment. How many would actually copy first few codes? Like the first and second atleast?


#9

@atulshanbhag do you mean to say people dont copy first and second problem at all??


#10

i dont say that, but MOSS does not catch everyone who does cheat as well. I would suggest the first few questions which are simple enough to be ignored by the plagiarism software


#11

Hey atulshanbhag, can you send an email to feedback@codechef.com with all the details, we will once again look into your case and help you with it.


#12

@admin, my ratings were restored. Thanks for resolving the issue.


#13

Sorry to be hard on the admins.