BARBER PRACTICE STATE BOARD
COMPREHENSIVE STUDY GUIDE 2026
FULL QUESTIONS AND SOLUTIONS
GRADED A+
◍ Steps of Louvain algorithm.
Answer: Step 0: each mode is its own communityStep 1: Repeat the
following, make biggest modularity increase by moving node from its
community to an adjacent node's community until no move increases
modularityStep 2: Each community is a super node, repeat step 1 using
super nodes, super arcs are created between super nodesIt is heuristic
◍ Collective outlier.
Answer: A set of data points that is (uncommonly) different from others -
for example, a missing heartbeat in an electrocardiogram
◍ What is modularity?.
Answer: measures of how well the graph is separated into communities or
modules that are connected a lot internally but not connected much in
between each other
◍ What are the three levels of neurons.
Answer: 1. Input level - external inputs, each neuron accepts a single piece
of information2. Hidden level3. Output level
◍ The flow of neural network.
Answer: each neuron gets an input from previous layer, calculates function
of weighted inputs, gives its output in next layer
◍ When we create models, we assume the system does not react; give
examples of when it does.
, Answer: Pricing decisions - company may model based on competitor
pricing, but the competitor pricing may change based on the company's
pricingGovernment tax policy - government set corporate tax policies and
companies react by how to store and spend money
◍ Two types of game theory.
Answer: 1.) competitive decision-making2.) cooperative game theory
◍ Concave function.
Answer: A function where the values are always above [or equal to] the
function's Line between end points.
◍ Timing of game theory.
Answer: 1.) making decisions simultaneously - one decision is made, can't
switch2.) Sequential - you make decisions sequentially, sometimes you can
see other side's strategy before making decision
◍ Types of strategy in game theory.
Answer: 1.) pure strategy - just one choice2.) mixed strategy - randomize
decisions according to probabilities; ex: rock, paper, scissors - if pure
strategy, like always picking rock, opponent will chose paper
◍ Information level of game theory.
Answer: 1.) perfection information - know all about everyone else's situation
(chess)2.) imperfect information - some have more information than others -
not symmetric (ex: gas station example)
◍ Concordance index.
Answer: Area under the ROC curve; an estimate of the classification
model's accuracy. Also called AUC.
◍ Confusion matrix.
Answer: Visualization of classification model performance.
◍ Zero sum game.
Answer: whatever one side gets, the other side loses
◍ Non-zero sum game.
, Answer: total benefit might be higher or lower
◍ What is prisoner's dilemma?.
Answer: in a game, even if players agree on the same price, both sides have
incentive to cheat, so both end up with lower price and everyone loses
◍ Contextual outlier.
Answer: A data point that is (uncommonly) far from other data points
related to it
◍ Convex function.
Answer: A function where the values are always below [or equal to] the
function's Line between end points..
◍ Corrected AIC.
Answer: Improved version of AIC, especially when sample size is small.
◍ Why is NLP hard?.
Answer: neural networks can recognize words, knowing what they mean
requires a different understanding
◍ What is NLP?.
Answer: general term of models of all type that try to help automat
understanding of language: speech to text, recognizing named entities,
determining meaning in context, understanding sentiment, choosing words
to convey message
◍ Correlation.
Answer: Relationship in which two things are likely to happen together,
regardless of whether one causes the other.
◍ Covariate.
Answer: A characteristic or measurement that can be used to estimate the
value of something.
◍ Cross-validation.
Answer: Validation technique where a model is tested on data different from
what it was trained on.
COMPREHENSIVE STUDY GUIDE 2026
FULL QUESTIONS AND SOLUTIONS
GRADED A+
◍ Steps of Louvain algorithm.
Answer: Step 0: each mode is its own communityStep 1: Repeat the
following, make biggest modularity increase by moving node from its
community to an adjacent node's community until no move increases
modularityStep 2: Each community is a super node, repeat step 1 using
super nodes, super arcs are created between super nodesIt is heuristic
◍ Collective outlier.
Answer: A set of data points that is (uncommonly) different from others -
for example, a missing heartbeat in an electrocardiogram
◍ What is modularity?.
Answer: measures of how well the graph is separated into communities or
modules that are connected a lot internally but not connected much in
between each other
◍ What are the three levels of neurons.
Answer: 1. Input level - external inputs, each neuron accepts a single piece
of information2. Hidden level3. Output level
◍ The flow of neural network.
Answer: each neuron gets an input from previous layer, calculates function
of weighted inputs, gives its output in next layer
◍ When we create models, we assume the system does not react; give
examples of when it does.
, Answer: Pricing decisions - company may model based on competitor
pricing, but the competitor pricing may change based on the company's
pricingGovernment tax policy - government set corporate tax policies and
companies react by how to store and spend money
◍ Two types of game theory.
Answer: 1.) competitive decision-making2.) cooperative game theory
◍ Concave function.
Answer: A function where the values are always above [or equal to] the
function's Line between end points.
◍ Timing of game theory.
Answer: 1.) making decisions simultaneously - one decision is made, can't
switch2.) Sequential - you make decisions sequentially, sometimes you can
see other side's strategy before making decision
◍ Types of strategy in game theory.
Answer: 1.) pure strategy - just one choice2.) mixed strategy - randomize
decisions according to probabilities; ex: rock, paper, scissors - if pure
strategy, like always picking rock, opponent will chose paper
◍ Information level of game theory.
Answer: 1.) perfection information - know all about everyone else's situation
(chess)2.) imperfect information - some have more information than others -
not symmetric (ex: gas station example)
◍ Concordance index.
Answer: Area under the ROC curve; an estimate of the classification
model's accuracy. Also called AUC.
◍ Confusion matrix.
Answer: Visualization of classification model performance.
◍ Zero sum game.
Answer: whatever one side gets, the other side loses
◍ Non-zero sum game.
, Answer: total benefit might be higher or lower
◍ What is prisoner's dilemma?.
Answer: in a game, even if players agree on the same price, both sides have
incentive to cheat, so both end up with lower price and everyone loses
◍ Contextual outlier.
Answer: A data point that is (uncommonly) far from other data points
related to it
◍ Convex function.
Answer: A function where the values are always below [or equal to] the
function's Line between end points..
◍ Corrected AIC.
Answer: Improved version of AIC, especially when sample size is small.
◍ Why is NLP hard?.
Answer: neural networks can recognize words, knowing what they mean
requires a different understanding
◍ What is NLP?.
Answer: general term of models of all type that try to help automat
understanding of language: speech to text, recognizing named entities,
determining meaning in context, understanding sentiment, choosing words
to convey message
◍ Correlation.
Answer: Relationship in which two things are likely to happen together,
regardless of whether one causes the other.
◍ Covariate.
Answer: A characteristic or measurement that can be used to estimate the
value of something.
◍ Cross-validation.
Answer: Validation technique where a model is tested on data different from
what it was trained on.
