README 1) Introduction 2) Directory structure 3) Instance structure 3) How the data was created 4) How the data can be used ***************************** 1) Introduction: This data corresponds to the data and experiemnts described in Section 5 of the following paper submitted to SEA conference 2018: A 3/2-approximation algorithm for the Student-Project Allocation problem Authors: Frances Cooper and David Manlove The data is located at: http://dx.doi.org/10.5525/gla.researchdata.583 The software is located at: https://doi.org/10.5281/zenodo.1183222 ***************************** 2) Directory structure: An 'Evaluations' directory containing the following sub-directories: - SIZE1, SIZE2, ..., SIZE10 - TIES1, TIES2, ..., TIES11 - PREF1, PREF2, ..., PREF10 - SCAL1, SCAL2, ..., SCAL5 - SCALP1, SCALP2, ..., SCALP6 Each of these sub-directories itself contains the following 5 sub-directories and one text file. - Instances - contains the randomly generated instances - ResultsApprox - approximation algorithm matching of each instance - ResultsOptimalMax - maximum stable matching for each instance - ResultsOptimalMin - minimum stable matching for each instance - Correctness - correctness testing results over above three matchings - info.text - instance generation parameter information ***************************** 3) Instance structure: : ... ... : : : ... : : : : ... ... Notes: In students preference lists, brackets '()' surround all tied projects. In lecturer preference lists, brackets '()' surround all tied students. ***************************** 3) How the data was created: Data creation is discussed in the paper. ***************************** 4) How the data can be used: This data can be used with the corresponding 3/2 approximation algorithm and IP implementations. Details of where to find this software can again be found in the paper.