The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

Latent Dirichlet Allocation coupled with Bayesian Time Series analyses

R-CMD-check License Lifecycle:maturing Codecov test coverage CRAN downloads DOI

Overview

The LDATS package provides functionality for analyzing time series of high-dimensional data using a two-stage approach comprised of Latent Dirichlet Allocation (LDA) and Bayesian time series (TS) analyses.

For a full description of the math underlying the LDATS package, see the technical document.

Status: Stable Version Available, Continuing Development

A stable version of LDATS is available on CRAN, but the package is actively being developed by the Weecology Team. The API is well defined at this point and should not change substantially.

Installation

You can install the stable version of LDATS from CRAN with:

To obtain the current development version of LDATS from GitHub, install the devtools package and then use it to install LDATS:

install.packages("devtools")
devtools::install_github("weecology/LDATS")

Usage

Here is an example of a full LDA-TS analysis using the Portal rodent data:

library(LDATS)
data(rodents)
r_LDATS <- LDA_TS(rodents, topics = 2:5, nseeds = 2, formulas = ~1,  
                  nchangepoints = 0:1, timename = "newmoon")

Which conducts two replicates (nseeds) for each of two to five topics in an LDA model using the document term table, selects the best (AIC) of those, then conducts two time series models on it (an intercept-only model under 0 and 1 changepoints), then selects the best (AIC) of the time series, and packages all the models together. This uses the document term table to weight the samples by their sizes (number of words) and instructs the function to use the column named "newmoon" in the document covariates table as the time variable.

The resulting object is of class LDA_TS, which has a few basic routines available:

print(r_LDATS)

prints the selected LDA and TS models and

plot(r_LDATS)

produces a 4-panel figure of them a la Figure 1 from Christensen et al. 2018.

More Information

Based on initial work using LDA to analyze time-series data at Portal by Erica M. Christensen, David J. Harris, and S. K. Morgan Ernest, which has been published in Ecology

Acknowledgements

The motivating study—the Portal Project—has been funded nearly continuously since 1977 by the National Science Foundation, most recently by DEB-1622425 to S. K. M. Ernest, which also supported (in part) E. Christensen’s time. Much of the computational work (including time of J. Simonis, D. Harris, and H. Ye) was supported by the Gordon and Betty Moore Foundation’s Data-Driven Discovery Initiative through Grant GBMF4563 to E. P. White. R. Diaz was supported in part by a National Science Foundation Graduate Research Fellowship (No. DGE-1315138 and DGE-1842473).

Author Contributions

J. L. Simonis provided insight on LDA applications and feedback on technical writing during development of the first version of the LDATS model and application, led the coding and mathematical development of the model into an R package, and led writing on the technical model document. E. M. Christensen led the project during development of the first version of the LDATS model and its application to the Portal data, specifically conceiving the project, coding the pipeline wrappers of the analysis, and writing and editing the first description of the model and its application (Christensen et al. 2018). D. J. Harris was involved in developing and applying the first version of the LDATS model, specifically suggesting the LDA and change point approaches, coding the first version of the change point model, and writing and editing the first description of the model (Christensen et al. 2018). R. Diaz contributed code to the LDATS package, wrote vignettes, provided insight into model development, and conducted extensive end-user code application testing. H. Ye contributed code to the LDATS package, insight into data structures and LDA algorithms, and significant feedback on vignettes. E. P. White helped design, troubleshoot, and supervise initial methods development; provided big-picture feedback on development of the R package; contributed end-user application testing; and gave substantial editing feedback on the technical document. S. K. Morgan Ernest provided managerial oversight and feedback on the project in both the initial and second stages of LDATS development, tested applications of the code to data sets, and assisted with writing and editing of the first description of the model and its application (Christensen et al. 2018) as well as the technical model document.

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.