diff --git a/.Rbuildignore b/.Rbuildignore
index 91114bf..96ece55 100644
--- a/.Rbuildignore
+++ b/.Rbuildignore
@@ -1,2 +1,5 @@
 ^.*\.Rproj$
 ^\.Rproj\.user$
+^\.travis\.yml$
+^data-raw$
+.gitlab-ci.yml
diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
new file mode 100644
index 0000000..4d8aab9
--- /dev/null
+++ b/.gitlab-ci.yml
@@ -0,0 +1,41 @@
+document:
+  stage: document
+  script:
+  - R --vanilla -e 'library(devtools);  devtools::document(); devtools::document()'
+  artifacts:
+    paths:
+    - man/
+    - NAMESPACE
+
+test:
+  stage: test
+  script:
+  - R --vanilla -e 'library(devtools); library(testthat); devtools::test(reporter
+    = StopReporter)'
+  dependencies:
+  - document
+
+build:
+  stage: build
+  script:
+  - R --vanilla -e 'library(devtools); devtools::build(path = "./")'
+  artifacts:
+    paths:
+    - '*.tar.gz'
+  dependencies:
+  - document
+
+check:
+  stage: check
+  script:
+  - R --vanilla -e 'library(devtools);  tar_file <- file.path(getwd(), list.files(".",
+    pattern = ".tar.gz")); results <- devtools::check_built(tar_file); stopifnot(sum(length(results$errors), length(results$warnings)) <= 0)'
+  dependencies:
+  - build
+
+stages:
+- document
+- test
+- build
+- check
+
diff --git a/.travis.yml b/.travis.yml
new file mode 100644
index 0000000..8d139ac
--- /dev/null
+++ b/.travis.yml
@@ -0,0 +1,5 @@
+# R for travis: see documentation at https://docs.travis-ci.com/user/languages/r
+
+language: R
+sudo: false
+cache: packages
diff --git a/NAMESPACE b/NAMESPACE
index c047479..f652966 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -4,4 +4,6 @@ export(biased_trans_matrix)
 export(cbrw)
 import(dplyr)
 import(rlang)
+importFrom(stats,setNames)
 importFrom(tidyr,spread)
+importFrom(utils,combn)
diff --git a/R/cbrw.R b/R/cbrw.R
index 11b5aff..3b2b6c5 100644
--- a/R/cbrw.R
+++ b/R/cbrw.R
@@ -1,7 +1,7 @@
 #' @title Coupled Biased Random Walks  
-#'  TODO
+#' @description TODO
 #' @param data a data.frame containing catgorical data
-#' @value the input data frame with an additional \emph{score} variable representing
+#' @return the input data frame with an additional \emph{score} variable representing
 #'        relative outlier-ness of the observation
 #' @export
 cbrw <- function(data) {
diff --git a/R/count.R b/R/count.R
index e8fdff0..bf317aa 100644
--- a/R/count.R
+++ b/R/count.R
@@ -1,11 +1,11 @@
 # TODO: Probably consolidate, or just build an R6 class
-#' @title Counting helper functions
-#' @name counting_helpers
+# @title Counting helper functions
+# @name counting_helpers
 
-#' @rdname counting_helpers
-#' @param data a tibble of categorical data
-#' @return a tibble containing all unique combinations of feature values
-#'    with columns: \emph{u}, \emph{v}, \emph{freq}, emph{p}, and \emph{feature}
+# @rdname counting_helpers
+# @param data a tibble of categorical data
+# @return a tibble containing all unique combinations of feature values
+#    with columns: \emph{u}, \emph{v}, \emph{freq}, emph{p}, and \emph{feature}
 intra_freq <- function(data) {
   var_quos <- lapply(names(data), as.name)
 
@@ -30,12 +30,14 @@ intra_freq <- function(data) {
   return(out)
 }
 
-#' Helper function to calculate all bivariate frequencies
-#' within the dataset
-#' @param data a tibble of categorical data
-#' @return a tibble containing all unique combinations of feature values
-#'    with columns: \emph{u}, \emph{v}, \emph{freq}, and \emph{group}
-#' @rdname counting_helpers
+# Helper function to calculate all bivariate frequencies
+# within the dataset
+# @param data a tibble of categorical data
+# @return a tibble containing all unique combinations of feature values
+#    with columns: \emph{u}, \emph{v}, \emph{freq}, and \emph{group}
+# @rdname counting_helpers
+#' @importFrom utils combn
+#' @importFrom stats setNames
 inter_freq <- function(data) {
 
   # Calculate all 2 variable combinations of variables
diff --git a/R/data.R b/R/data.R
index fdb3617..e703df1 100644
--- a/R/data.R
+++ b/R/data.R
@@ -1,3 +1,4 @@
 #' @title CBRW paper exmaple dataset
+#' @description See dataset in cited paper (TODO)
 #' Example dataset as shown on the first page of the paper
 "cbrw_example"
\ No newline at end of file
diff --git a/R/matrix.R b/R/matrix.R
index 9ac6ecc..e0d02f3 100644
--- a/R/matrix.R
+++ b/R/matrix.R
@@ -1,4 +1,8 @@
 #' @title Generate Biased Transition Matrix
+#' @description
+#' Takes in a data.frame of categorical data and returns a weighted transition matrix
+#' which characterizes the edge weights of a directed graph representation of the inter-feature value couplings
+#'
 #' @param data a data.frame containing mosly categorical data
 #' @param all_data a boolean (default: FALSE) on whether to return additional node and edge data (see Note)
 #' @return a biased transition matrix of dim [k,k] where k is the number of unique feature values
diff --git a/README.md b/README.md
index 5567dc8..9daa768 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,4 @@
-# cbRw, Coupled Biased Random Walks  
+# cbRw, Coupled Biased Random Walks  [![Build Status](https://travis-ci.org/beansrowning/cbRw.svg?branch=master)](https://travis-ci.org/beansrowning/cbRw)  
 Anomaly detection for complex categorical data  
 
 ## Overview  
@@ -7,7 +7,15 @@ Described by [Pang, Cao, and Chen (2016)](https://www.ijcai.org/Proceedings/16/P
 Also based on work by Daniel Kaslovsky in the Python implementation, [Coupled-Biased-Random-Walks
 ](https://github.com/dkaslovsky/Coupled-Biased-Random-Walks)
 
-## Work in progress
+## Installation  
+
+*Note that this is still very much developmental*  
+
+```r
+# Installation is straightforward with devtools
+# install.packages("devtools")
+devtools::install_github("beansrowning/cbRw")
+```  
 
 ## Public Domain
 This repository constitutes a work of the United States Government and is not
diff --git a/man/biased_trans_matrix.Rd b/man/biased_trans_matrix.Rd
index 6673df2..e7f0732 100644
--- a/man/biased_trans_matrix.Rd
+++ b/man/biased_trans_matrix.Rd
@@ -14,6 +14,10 @@ biased_trans_matrix(data, all_data = FALSE)
 \value{
 a biased transition matrix of dim [k,k] where k is the number of unique feature values
 }
+\description{
+Takes in a data.frame of categorical data and returns a weighted transition matrix
+which characterizes the edge weights of a directed graph representation of the inter-feature value couplings
+}
 \section{Note}{
 
 If all_data is `TRUE`, A list with three members will be returned:
diff --git a/man/cbrw.Rd b/man/cbrw.Rd
new file mode 100644
index 0000000..9c645e9
--- /dev/null
+++ b/man/cbrw.Rd
@@ -0,0 +1,18 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/cbrw.R
+\name{cbrw}
+\alias{cbrw}
+\title{Coupled Biased Random Walks}
+\usage{
+cbrw(data)
+}
+\arguments{
+\item{data}{a data.frame containing catgorical data}
+}
+\value{
+the input data frame with an additional \emph{score} variable representing
+       relative outlier-ness of the observation
+}
+\description{
+TODO
+}
diff --git a/man/cbrw_example.Rd b/man/cbrw_example.Rd
index c26f2a8..e4544a7 100644
--- a/man/cbrw_example.Rd
+++ b/man/cbrw_example.Rd
@@ -3,10 +3,13 @@
 \docType{data}
 \name{cbrw_example}
 \alias{cbrw_example}
-\title{CBRW paper exmaple dataset
-Example dataset as shown on the first page of the paper}
+\title{CBRW paper exmaple dataset}
 \format{An object of class \code{spec_tbl_df} (inherits from \code{tbl_df}, \code{tbl}, \code{data.frame}) with 12 rows and 5 columns.}
 \usage{
 cbrw_example
 }
+\description{
+See dataset in cited paper (TODO)
+Example dataset as shown on the first page of the paper
+}
 \keyword{datasets}
diff --git a/man/counting_helpers.Rd b/man/counting_helpers.Rd
deleted file mode 100644
index 629176c..0000000
--- a/man/counting_helpers.Rd
+++ /dev/null
@@ -1,28 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/count.R
-\name{counting_helpers}
-\alias{counting_helpers}
-\alias{intra_freq}
-\alias{inter_freq}
-\title{Counting helper functions}
-\usage{
-intra_freq(data)
-
-inter_freq(data)
-}
-\arguments{
-\item{data}{a tibble of categorical data}
-
-\item{data}{a tibble of categorical data}
-}
-\value{
-a tibble containing all unique combinations of feature values
-   with columns: \emph{u}, \emph{v}, \emph{freq}, emph{p}, and \emph{feature}
-
-a tibble containing all unique combinations of feature values
-   with columns: \emph{u}, \emph{v}, \emph{freq}, and \emph{group}
-}
-\description{
-Helper function to calculate all bivariate frequencies
-within the dataset
-}
diff --git a/tests/testthat.R b/tests/testthat.R
new file mode 100644
index 0000000..58e1382
--- /dev/null
+++ b/tests/testthat.R
@@ -0,0 +1,4 @@
+library(testthat)
+library(cbRw)
+
+test_check("cbRw")
diff --git a/tests/testthat/test_cbrw.R b/tests/testthat/test_cbrw.R
new file mode 100644
index 0000000..94753ef
--- /dev/null
+++ b/tests/testthat/test_cbrw.R
@@ -0,0 +1,27 @@
+context("Algorithm accuracy")
+
+test_that("Score values are equal to canonical", {
+
+  # Canonical values pulled from running cbrw test data through
+  # the python CBRW package, dkaslovsky/Coupled-Biased-Random-Walks
+  canonical <- c(
+    0.123556072282254,
+    0.0504085179712585,
+    0.0476111020550283,
+    0.0495085666704524,
+    0.0844538406306188,
+    0.0468666953948249,
+    0.0476111020550283,
+    0.0493436236504131,
+    0.0459772220233669,
+    0.0694800337106498,
+    0.0495038674807341,
+    0.0668533395540888
+  )
+
+  # Process example dataset and compare values
+  data(cbrw_example)
+  cbrw_example <- cbrw(cbrw_example)
+
+  expect_equal(cbrw_example$score, canonical)
+})
\ No newline at end of file