IAQ index and updates
j	surrogate parameter
Φ j *	fractional dose
Φj	exposure level
Φj,e	reference exposure limit
θ	IAQ index
r	relative impact ratio
xu/xs	distribution functions for unsatisfactory/satisfactory IAQ index
LR	likelihood ratio
Data processing	Data processing
X	data vector
rd/1 − rd	test data/training data
nd,t/nd,g	number of data points in the test/training set
AC	model accuracy
ACbl	baseline accuracy
TP/TN	true positive/negative
FP/FN	false positive/negative
N	sample size
K	number of folds
Units for IAQ parameters
ppm	parts per million
μg m−3	microgram per cubic meter
Bq m−3	becquerels per cubic meter
CFU m−3	colony-forming units per cubic meter
Regularization
f	cost function
yi	true value
x β	predicted value
C	regularization factor
n	number of dimensions
Decision tree/random forest
pj2	probability of j
j	class
D	tree’s maximum depth
ns/nr	minimum number of samples required to split an internal node/be at a leaf node
nf	number of trees
Support Vector Machines
α, β	constants
xi	inputs
yi	output class
M	margin half-width
εi	slack variables
c0, c1	hyperparameters for K(xi,xj)
K(xi,xj)	kernel function
γ	kernel coefficient
k-Nearest Neighbors
k	constant
d(xi,yi)	Euclidean distance
$\widehat{y}$	predictions
W	weight function
dk−1	neighbour distance
MLP-ANN
R	dataset
m/o	dimension for input/output
J	local gradient of function f
β	parameter
y	independent variables
δ	increment
Logistic regression
x 0	sigmoid’s midpoint of x
x	inputs
k	logistic growth rate
w	coefficient vector