absFromDataframe: Generate pairwise absolute differences from dataframes

In xhtai/heisenbrgr: Linking Accounts on Anonymous Marketplaces

Description

Given a dataframe where each row corresponds to a vendor hash (can be generated using the functions in 'step2_anytime.R'), generate absolute differences between columns. Columns need to be numeric variables. There is also the option to set the result to -1 if the variable being compared has value 0 for both accounts in the pair. This is useful for variables such as the number of tokens in the profiles – if both have no profile information the similarity should be missing (-1) instead of 0.

Usage

absFromDataframe(pairsOfInterest, dtfName, varName, setToMinus1)

Arguments

pairsOfInterest	dataframe with columns 'hash1', 'hash2' referring to vendor hashes for the comparison
dtfName	dataframe containing the columns to be compared
varName	vector of variable names that we want to compute differences for (TODO: now only works if >1 col)
setToMinus1	logical vector, same length as 'varName'

Value

vector the length of 'nrow(pairsOfInterest)', containing the resulting similarity scores for each pair

xhtai/heisenbrgr documentation built on June 8, 2019, 9:30 a.m.