Bayesian Preference Learning with the Mallows Rank Model


[Up] [Top]

Documentation for package ‘BayesMallows’ version 1.1.2

Help Pages

assess_convergence Trace Plots from Metropolis-Hastings Algorithm
assign_cluster Assign Assessors to Clusters
BayesMallows BayesMallows: Bayesian Preference Learning with the Mallows Rank Model.
beach_preferences Beach Preferences
calculate_backward_probability Calculate Backward Probability
calculate_forward_probability Calculate Forward Probability
compute_consensus Compute Consensus Ranking
compute_consensus.BayesMallows Compute Consensus Ranking
compute_consensus.consensus_SMCMallows Compute Consensus Ranking
compute_mallows Preference Learning with the Mallows Rank Model
compute_mallows_mixtures Compute Mixtures of Mallows Models
compute_posterior_intervals Compute Posterior Intervals
compute_posterior_intervals.BayesMallows Compute posterior intervals
compute_posterior_intervals.SMCMallows Compute posterior intervals
compute_posterior_intervals_alpha Compute Posterior Intervals Alpha
compute_posterior_intervals_rho Compute Posterior Intervals Rho
compute_rho_consensus Compute rho consensus
correction_kernel Correction Kernel
correction_kernel_pseudo Correction Kernel (pseudolikelihood)
create_ordering Convert between ranking and ordering.
create_ranking Convert between ranking and ordering.
estimate_partition_function Estimate Partition Function
expected_dist Expected value of metrics under a Mallows rank model
generate_constraints Generate Constraint Set from Pairwise Comparisons
generate_initial_ranking Generate Initial Ranking
generate_transitive_closure Generate Transitive Closure
get_mallows_loglik Get Mallows log-likelihood
get_sample_probabilities Get Sample Probabilities
label_switching Checking for Label Switching in the Mallows Mixture Model
leap_and_shift_probs Leap and Shift Probabilities
lik_db_mix Likelihood and log-likelihood evaluation for a Mallows mixture model
metropolis_hastings_alpha Metropolis-Hastings Alpha
metropolis_hastings_aug_ranking Metropolis-Hastings Augmented Ranking
metropolis_hastings_aug_ranking_pseudo Metropolis-Hastings Augmented Ranking (pseudolikelihood)
metropolis_hastings_rho Metropolis-Hastings Rho
obs_freq Observation frequencies in the Bayesian Mallows model
plot.BayesMallows Plot Posterior Distributions
plot_alpha_posterior Plot Alpha Posterior
plot_elbow Plot Within-Cluster Sum of Distances
plot_rho_posterior Plot the posterior for rho for each item
plot_top_k Plot Top-k Rankings with Pairwise Preferences
potato_true_ranking True ranking of the weights of 20 potatoes.
potato_visual Result of ranking potatoes by weight, where the assessors were only allowed to inspected the potatoes visually. 12 assessors ranked 20 potatoes.
potato_weighing Result of ranking potatoes by weight, where the assessors were allowed to lift the potatoes. 12 assessors ranked 20 potatoes.
predict_top_k Predict Top-k Rankings with Pairwise Preferences
print.BayesMallows Print Method for BayesMallows Objects
print.BayesMallowsMixtures Print Method for BayesMallowsMixtures Objects
rank_conversion Convert between ranking and ordering.
rank_distance Distance between a set of rankings and a given rank sequence
rank_freq_distr Frequency distribution of the ranking sequences
sample_dataset A synthetic 3D matrix ('n_users', 'n_items', 'Time') generated using the sample_mallows function. These are test datasets used to run the SMC-Mallows framework for the cases where we know all of the users in our system and their original ranking information are partial rankings. However at some point in time, we observe extra information about an existing user in the form of a rank for an item that was previously not known ('NA'). These datasets are very contrived as the first time step ('sample_dataset[, , 1]') we observed the top 'm / 2' items from each user, where 'm' is the number of items in a ranking. Then, as we increase the time, we observe the next top ranked item from one user at a time, then the next top ranked item, and so on until we have a complete dataset at 'sample_dataset[, , Time]'.
sample_mallows Random Samples from the Mallows Rank Model
smc_mallows_new_item_rank SMC-Mallows new users rank
smc_mallows_new_item_rank_alpha_fixed SMC-Mallows new item rank (alpha fixed)
smc_mallows_new_users_complete SMC-Mallows New Users Complete
smc_mallows_new_users_partial SMC-Mallows new users partial
smc_mallows_new_users_partial_alpha_fixed SMC-mallows new users partial (alpha fixed)
smc_processing SMC Processing
sushi_rankings Sushi Rankings