Part 1 Introduction

1.1 What is cookiemilk

The Python package cookiemilk is an easy-to-use tool to process knowledge structure data automatically and quickly. It includes (a) data conversion, such as converting concept maps or essays to graphs, (b) graph-based features calculation, and (c) graph visualization. The cookiemilk package is primarily based on the Linear Aggregate Approach (Clariana et al., 2009) but also allows for the calculation of some features introduced by Pirnay-Dummer and Ifenthaler (2011).

The code was initially written for a specific analysis purpose when I was working on my Master’s thesis. Although I reorganized the code to make it flexible for use in different analysis plans, the package has not been fully validated, and some errors may occur. Please feel free to contact me via weiziqianpsych@outlook.com to report any issues and help me improving the package and this tutorial website!

The name of the package is inspired by my cats, Cookie (the grey-strip one in the figure below) and Milk (the black one in the figure below).

References

Clariana, R. B., Wallace, P. E., & Godshalk, V. M. (2009). Deriving and measuring group knowledge structure from essays: The effects of anaphoric reference. Educational Technology Research and Development, 57, 725-737.

Pirnay-Dummer, P., & Ifenthaler, D. (2011). Reading guided by automated graphical representations: How model-based text visualizations facilitate learning in reading comprehension tasks. Instructional Science, 39(6), 901-919.

1.2 Installation

Windows:

py -m pip install --upgrade cookiemilk

Unix/macOS:

python3 -m pip install --upgrade cookiemilk

NOTE: Like many other Python packages, cookiemilk functions the same way on devices with different operating systems. This means you can run the example code in this tutorial on either a Windows PC or a Mac. The only difference is the method of installation.

1.3 Import

import cookiemilk

1.4 Main functions

Usage	Functions	Description
Data arrangement	cmap2graph	import a concept map and convert it into a graph
	text2graph	import a text and convert it into a graph
Data processing	pathfinder_network	calculate the PFNet of a given graph
	calc_surface_matching	calculate the surface similarity between two graphs
	calc_graphical_matching	calculate the graphical similarity between two graphs
	calc_gcent	calculate the graphical centrality (GC) of a given graph
	numerical_sim	calculate the numerical similarity
	calc_tversky	calculate Tversky’s similarity between two graphs
Visualization	draw	plot and/or save a graph
	average_graph	generate an average graph
Others	quick_analysis	conduct data analysis using the above functions

1.5 Dependencies

networkx: A Python package for network analysis

numpy: A Python package for scientific computing

pywebview: A Python package for building GUI with JavaScript, HTML, and CSS

d3.js: A JavaScript library for manipulating documents based on data

1.6 Citation

To cite cookiemilk in publications use:

Wei, Z., Zhang, Y., Clariana, R. B., & Chen, X. (2024). The effects of reading prompts and post-reading generative learning tasks on multiple document integration: Evidence from concept network analysis. Educational Technology Research and Development, 72, 661–685. https://doi.org/10.1007/s11423-023-10326-w