~bendersteed/markov-chains-web-navigation

b0516676049d5346fe3e9ec34760b82ff183e838 — Dimos Dimakakos 1 year, 9 months ago 95d4508 master
Docs: update README
1 files changed, 10 insertions(+), 2 deletions(-)

M README.md
M README.md => README.md +10 -2
@@ 1,9 1,10 @@
# Modeling web navigation data into Markov chains of varying order
The only dependency of the tool is the [tidyverse](https://www.tidyverse.org/) library.

This tool has the main entry point simulation:

``` R
simulation <- function(input, k, states, topics) { ... }
simulation <- function(input, k, states, topics, skip) { ... }
```

1. input: a file of navigation data, where each line is a navigation path of numbers that


@@ 11,8 12,15 @@ simulation <- function(input, k, states, topics) { ... }
2. k: the upper limit of Markov chain order, for which to model the data into
3. states: the number of states (topics) that appear in the dataset
4. topics: an array of length states, that maps index to name
5. skip: the number of lines to ignore from input

It returns 
Topics and skip arguments provide default values for the analysis of the msnbc anonymous navigation
dataset that can be found [here](http://kdd.ics.uci.edu/databases/msnbc/msnbc.html).

It returns three tibbles:
1. frequencies of topics
2. loglikelihood of topics
3. results that contain loglikelihood ratio statistics, AIC and BIC

There are also some helper functions provided for the creation of graphs.