next_release.md

In USCCANA/netdiffuseR: Analysis of Diffusion and Contagion Processes on Networks

Development road

The following document details expectations for the comming versions. There is no guarantee on the implementation of any of the points here.

CRAN 1.16.4: Expected

Polishing

• Review manuals: Description of arguments and name of the functions.
• ~~Aliases creation: select_egoalter table_...~~
• New meta on the diffnet class:
  • ~~name: 'A diffnet network'~~
  • ~~description 'Diffnet network with no description'~~
  • timestamp (maybe auto update) 'DATE'
  • author: 'user'
  • netdiffuseR.version: 1.16.6 Plus some other stuff:
  • coordinates (xyz)
  • colors
  • vertex size

Testing

• survey_to_diffnet, ~~edgelist_to_adjmat~~
• ~~diffnet_as_igraph, igraph_as_diffnet~~

Developing

• ~~diffnet_to_networkDynamic, networkDynamic_to_diffnet. Need to reach out the author of Carter Butts: Is there any way to access networkDynamic objects formal definition??~~

• ~~rewire_dgr_preserve: A brief comparison on igraph and netdiffuseR rewiring algorithms shows that igraph has no significan speed improvement in small-medium graphs. This may be due to having a similar rewiring algorithm. The dgr preserve should work in a similar fashion and should incorporate the call to std::<vector>remove method.~~

• ~~Raise the level of ARMA_BIT_64INT: Create a toy package in which, by simulations, we create big sparse matrices of class dgCMatrix do read/write operations both from R and from C++.~~

• read_ucinet and rename read_ucinet to read_dl. And write versions of it. It will be useful to ask Steven B. if is there any way of using UCINET to analyze dynamic data.

• read_net Talk with the developer to see is they have more information (aditional reference) on the data formatting for dynamic graphs and attributes. Recall UCINET's documents provided with spreadsheets pointing to data types and structure.

Improving

• ~~rdiffnet:~~ Set initial adopters by providing a vector:

  • logical
  • integer (of length n)
  • integer (of random length)
  • character (ids, rownames)

• What happens in survey_to_diffnet when a data.frame is provided, but se set to fill data and the checking done over the dimmensions of the data?

Examples

• Include the survival analysis example: Maybe use the recidivism data mixed with social data? Find a paper where someone does that
• Include a probit/logit model
• ~~Write down an example with the function select_egoalter and rename it.~~

USCCANA/netdiffuseR documentation built on Sept. 5, 2023, 12:31 a.m.