comparative-judgement 0.0.6

A package for conducting Comparative Judgement

Comparative Judgement

A package for comparative judgement (CJ).

Installation

Dependencies

comparative-judgement requires:

• Python (>= |PythonMinVersion|)
• NumPy (>= |NumPyMinVersion|)
• SciPy (>= |SciPyMinVersion|)
• Ray

User installation

If you already have a working installation of NumPy and SciPy, the easiest way to install comparative_judgement is using pip::

    pip install comparative-judgement

    conda install -c conda-forge comparative-judgement

Bayesian CJ

Importing the BCJ model and initiating a instance of the model with 4 samples:

from cj.models import BayesianCJ

BCJ = BayesianCJ(4)

Creating the data:

import numpy as np

data = np.asarray([
    [0, 1, 0],
    [0, 1, 0],
    [0, 3, 0],
    [1, 0, 1],
    [1, 0, 1],
    [1, 0, 1],
    [1, 2, 1],
    [1, 2, 1],
    [1, 2, 1],
    [1, 2, 1],
    [1, 2, 1],
    [2, 1, 2],
    [2, 1, 2],
    [2, 1, 2],
    [2, 3, 2],
    [3, 0, 3],
    [3, 0, 3],
    [3, 0, 3],
    [3, 0, 3],
    [3, 2, 3],
    [3, 2, 3],
    [3, 2, 3],
])

running the model:

BCJ.run(data)

Finding the $\mathbb{E}[\mathbf{r}]$

BCJ.Er_scores
>>> [3.046875, 2.09765625, 3.05859375, 1.796875]

Finding the BCJ rank:

BCJ.rank
>>> array([3, 1, 0, 2])

Multi-Criteria BCJ

Importing the BCJ model and initiating a instance of the model with 4 samples:

from cj.models import MBayesianCJ

criteria_weights = [0.2, 0.2, 0.6]

MBCJ = MBayesianCJ(3, criteria_weights)

data = [
    #A, B,C1, 2, 3  
    [0, 1, 1, 1, 1],
    [1, 2, 1, 1, 1],
    [0, 2, 0, 0, 2]
]

running the model:

MBCJ.run(data)

Finding the overall MBCJ rank:

MBCJ.combined_rank
>>> array([1, 2, 0])

Finding the individual criteria BCJ ranks:

MBCJ.lo_rank_scores
>>> {0: [np.float64(2.0), np.float64(1.5), np.float64(2.5)],
     1: [np.float64(2.0), np.float64(1.5), np.float64(2.5)],
     2: [np.float64(2.5), np.float64(1.5), np.float64(2.0)]}

Traditional BTM CJ

Importing the BTM Model a instance of the model with 4 samples:

from cj.models import BTMCJ

BTM = BTMCJ(4)

running the model:

BTM.run(data)

Finding the optimised p scores:

BTM.optimal_params
>>> array([-0.44654627,  0.04240265, -0.41580243,  0.81994508])

find BTM rank:

BTM.rank
>>> array([3, 1, 2, 0])

---

Pair Selection Methods

Entropy

from cj.pair_selector import EntropyPairSelector

entropy_pairs = EntropyPairSelector(5)

scores = [55, 65, 72, 45, 80]
standard_dev = 5
entropy_pairs.run_entropy_pairs_simulation(scores, standard_dev)

entropy_pairs.results
>>> [[1, 4, 4],
 [1, 3, 1],
 [2, 3, 2],
 [1, 2, 2],
 [0, 2, 2],
 [0, 4, 4],
 [0, 1, 1],
 [3, 4, 4],
 [0, 3, 0],
 [2, 4, 4],
 [1, 2, 2],
 [2, 3, 2],
 [1, 4, 4],
 [3, 4, 4],
 [2, 4, 4],
 [0, 4, 4],
 [0, 2, 2],
 [0, 1, 1],
 [0, 3, 0],
 [1, 3, 1],
 [1, 4, 4],
 [0, 1, 1],
 [0, 4, 4],
 [3, 4, 4],
 [2, 3, 2],
...
 [2, 4, 4],
 [0, 1, 1],
 [0, 4, 4],
 [2, 3, 2],
 [1, 2, 2]]

---

Citing this Library:

@misc{comparative_judgement,
    author = {Andy Gray},
    title = {Comparative Judgement},
    year = {2024},
    publisher = {Python Package Index (PyPI)},
    howpublished = {\url{https://pypi.org/project/comparative-judgement/}}
}