hoo.horizon.DT: Horizon of Observation Analysis for Multimodal Data,...

In scaotravis/hoo: Calculate adjacency vectors using horizon of observation model

Description

For multimodal data, players might not fully observe other player's actions. This function takes such factor into account and return the adjacency matrix for the actions players can observe. Specifically, this function utilizes data.table structure and lapply() function in attempt to increase performance.

Usage

hoo.horizon.DT(data, Units, Conversation, Codes, dataModeCol, modeObserve,
  usersCol, windowSize)

Arguments

data	Multimodal data.frame or data.table
Units	A vector of Strings describing the Units for ENA model
Conversation	A vector of Strings describing the Conversations took place in ENA model
Codes	A vector of Strings entailing the big C Codes of interest for ENA analysis
dataModeCol	Name of the column where the types of multimodal data is stored
modeObserve	Modes of data where actions are observable to all players
usersCol	Name of the column entailing the unique user tracking info
windowSize	Size of the moving stanza window, for looking backwards (for whole conversation, input 1)

Value

a data frame containing the adjacency vectors of each ENA Units within data

Examples

adj = hoo.horizon.DT(data = mock,
                     Units = c("site", "userName"),
                     Conversation = c("site"),
                     Codes = c("Code1", "Code2", "Code3", "Code4"),
                     dataModeCol = "data",
                     modeObserve = "chat",
                     usersCol = "userName",
                     windowSize = 4)

scaotravis/hoo documentation built on July 5, 2020, 4:42 p.m.