Skip to content

A solver for 'College Admission Problem with Consent' based on paper 'Legal Assignments and fast EADAM with consent via classical theory of stable matchings'.

License

Notifications You must be signed in to change notification settings

HE-Xinyu/fast-EADAM

Repository files navigation

fast-EADAM

A solver for college admission problem with consent based on paper 'Legal Assignments and fast EADAM with consent via classical theory of stable matchings'.

Warning: It is almost finished, but the correctness and efficiency are not guaranteed.

How to use

There are two modes available cuurently:

  1. Performance test mode. It tests how the algorithm performs in randomly generated large datasets. You will be prompted to enter parameters required. Although time complexity doesn't change, my implementation is hundreds times faster than the original one.
  2. Single problem solver. You will be prompted to enter input and output file paths. The fully tested algorithms are Gale-Shapley algorithm, fast EADAM and vNM stable set.

main.cpp contains several usage examples.

Please check the test folder for examples of input and output. The input file should contain information in order below:

  1. num_of_students (n), num_of_schools (m);
  2. m numbers of seats of schools;
  3. n numbers of students consenting or not (1: yes; 0: no);
  4. a n * m matrix representing preference of students;
  5. a m * n matrix representing preference of schools.

Current Progress

Build the basic data structure;

Implement Gale-Shapley Algorithm in O(E) time and space complexity;

Implememt fast EADAM Algorithm in O(E) time and space complexity.

Add comments in the code.

The fast EADAM can deal with situations when the number of students does not equal to the number of seats.

Build performance test module.

Add copy destructors to avoid memory overflow.

Use operator overloading to make the code clear.

Implement vNM stable set algorithm in O(E).

Implement some help functions for comparing solution sets.

Provide a script (result_to_xlsx.py) for transforming output into excel file.

Hint

It is highly likely that the queue can be implemented in arrays to speed up the algorithm even more.

But the time and space complexity remain unchanged.

License

MIT

Reference

For the origin paper that published the algorithm, please refer to https://arxiv.org/pdf/1809.08506.

About

A solver for 'College Admission Problem with Consent' based on paper 'Legal Assignments and fast EADAM with consent via classical theory of stable matchings'.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published