Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Change the current algorithm to select problem in the suggested practice section #8

Open
shrinish123 opened this issue Dec 3, 2022 · 6 comments

Comments

@shrinish123
Copy link
Owner

Right now the current algorithm just selects 100 points higher rated problems than your current rating. We are in the need to implement a better algorithm, Suggestion from your side are welcome .
One suggestion is to fetch problems that helped people to improve their rating and then as per user rating, suggest them to the user.

@MaYaNkKashyap681
Copy link
Contributor

can we add something that predicts the problems according to the weak topics of any user?

@shrinish123
Copy link
Owner Author

Yes maybe you can do it. But what algorithm and API endpoints will you be using in order to predict the weaker topics of user ?
You can work on it , if you are interested

@MaYaNkKashyap681
Copy link
Contributor

We are making a graph for the user, so by using that graph we can find out the section he has solved the least.

1). In which file do you have the logic for this algorithm?

@MaYaNkKashyap681
Copy link
Contributor

1). In which file do you have the logic for this algorithm?
In which file do i get this algorithm.

@shrinish123
Copy link
Owner Author

shrinish123 commented Dec 19, 2022

The graph is just from the data being fetched over the submissions API endpoint of the user. You can check the algorithm to make the graph in the Analysis Page component.
If you are talking about the suggested practice algorithm you can directly check it in the Suggested Practice Page component.
It just takes the current rating of the user and makes 3 levels for rating + 100 level problems, rating + 200 level problems, rating +300 level problems.
But there is an issue in your algorithm, where you find a person's weak topic by the number of questions he has solved. For example, I have solved many problems of DP because I am the weakest at it, but as per your algo it will show I am strong at it. which contradicts the fact. So we need to make some better way to look at it.
Maybe a probable algo can be to check no. of wrong submissions on every topic and where a particular person's accuracy is low, i.e. correctSubmssionCount / Total SubmissionCount per topic, to predict the topic strength for the user.
I guess this can be a thing to implement in the analysis Page. Maybe I will make an issue on this topic. You can work on it if you would want to.

@MaYaNkKashyap681
Copy link
Contributor

Yes I want to work on this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants