PyMCDM - Re-identify

Python 3 library for re-identification of multi-criteria models.

Documentation is avaliable on readthedocs.

Installation

You can download and install pymcdm-reidentify library using pip:

pip install pymcdm-reidentify

You can run all tests with following command from the root of the project:

python -m unittest -v

Available methods

The library contains:

• Re-identification methods:

• Normalization methods:

• COMET Tools

Usage example

Below is a simple example of the re-identification of a decision-making model. For more examples with explanation see examples.

import numpy as np
from mealpy.swarm_based.PSO import OriginalPSO
from pymcdm.methods import TOPSIS
from pymcdm_reidentify.methods import SITW

# Define exemplary data
# Decision matrix
matrix = np.random.random((1000, 2)) 
# Types of critieria
types = np.array([-1, 1]) # 
# Unknown expert criteria weights 
# For purpose of re-identifiaction method
weights = np.random.random((2))
weights = weights / np.sum(weights) 

# Define exemplary unknown expert model
preference = TOPSIS()(matrix, weights, types)
rank = TOPSIS().rank(preference)

# Create re-identifiaction object
stoch = OriginalPSO(epoch=1000, pop_size=100)
model = SITW(stoch.solve, TOPSIS(), types)

# Fit model
model.fit(matrix, rank, log_to=None)

References

[1] Kizielewicz, B., Więckowski, J., & Sałabun, W. (2024, June). SESP-SPOTIS: Advancing Stochastic Approach for Re-identifying MCDA Models. In International Conference on Computational Science (pp. 281-295). Cham: Springer Nature Switzerland.

[2] Kizielewicz, B. (2022). Towards the identification of continuous decisional model: the accuracy testing in the SITCOM approach. Procedia Computer Science, 207, 4390-4400.

[3] Kizielewicz, B., Paradowski, B., Więckowski, J., & Sałabun, W. (2022). Identification of weights in multi-cteria decision problems based on stochastic optimization.

[4] Kizielewicz, B., Więckowski, J., Paradowski, B., Shekhovtsov, A., Wątróbski, J., & Sałabun, W. (2024, April). Stochastic Approaches for Criteria Weight Identification in Multi-criteria Decision Analysis. In Asian Conference on Intelligent Information and Database Systems (pp. 40-51). Singapore: Springer Nature Singapore.

[5] Kizielewicz, B., & Dobryakova, L. (2023). Stochastic Triangular Fuzzy Number (S-TFN) Normalization: A New Approach for Nonmonotonic Normalization. Procedia Computer Science, 225, 4901-4911.

[6] Kizielewicz, B., Więckowski, J., Paradowski, B., & Sałabun, W. (2022, July). Dealing with nonmonotonic criteria in decision-making problems using fuzzy normalization. In International conference on intelligent and fuzzy systems (pp. 27-35). Cham: Springer International Publishing.

[7] Kizielewicz, B., Więckowski, J., & Jankowski, J. (2023, September). Towards Re-identification of Expert Models: MLP-COMET in the Evaluation of Bitcoin Networks. In Special Sessions in the Information Technology for Business and Society Track of the Conference on Computer Science and Intelligence Systems (pp. 3-22). Cham: Springer Nature Switzerland.