ThreadNet weaves threads into networks
ThreadNet is a tool for visualization of repetitive sequences, such as organizational routines. It emphasizes the role of sequential and temporal context. It is being created for NSF (SES-1734237) Antecedents of Complexity in Healthcare Routine, a collaborative project between Michigan State University and the University of Rochester Medical Center. Co-PIs: Brian Pentland and Kenneth Frank (MSU), Julie Ryan Wolf and Alice Pentland (URMC). The original version of ThreadNet was implemented in MatLab.
ThreadNet is currently not available through CRAN. You can install the package directly rom the source, using devtools
:
if (!"devtools" %in% installed.packages()[, "Package"]) {
install.packages("devtools")
}
devtools::install_github('ThreadNet/ThreadNet')
In order to start the app:
library(ThreadNet)
ThreadNet()
ThreadNet uses an intuitive R Shiny
graphical user interface that you can explore on your own. For further documentation and sample data see ThreadNet's institutional homepage.
ThreadNet reads data in simple .CSV format and .XES format (IEEE standard for process event log data).
When using timestamped data, the first column must be called "tStamp". The timestamps should be in default R format: "yyyy-mm-dd hh:mm:ss"
Alternatively, the first colum can contain ordinal sequences numbers (1, 2, 3...) for each thread. For sequence numbers, the first column must be called "sequence".
Think of each row in the .CSV as a moment in time. What contextual factors do you need to describe that moment? You can include as many contextual factors as you need. The data are treated as case-sensitive text.
- Use consistent labels
- Fill in all the values for every row.
- Spaces will be replaced with underscore