France: Exploring Historical Cookbooks

Research questions

1. What were the main ingredients used in 1900 in France?

2. Can we observe a difference per region?

Project Plan and Milestones

Date	Tasks	Completion
Week 3	Find multiple French cookbood in French or English from different times. Prepare slides for the initial project idea presentation.	✓
Week 4	Compare different cookbooks, consider the OCR scan and think of possible research questions. Discuss with TAs the goal and implementation of the projects.	✓
Week 5	Decide on one French cookbook. Scan the physical book.	✓
Week 6-7	Give OCR scan for the pages. Start to construct the dataset.	✓
Week 8-9	Prepare for midterm presentaton. Construct dataset and think of data structure to store the information.	✓
Week 10	Set up the GitHub repository. Finish the creation of the dataset	✓
Week 11	Fix bugs in extraction script and take exceptional cases into consideration. Create categories for ingredients.	✓
Week 12	Perform the data processing of the ingredients. Exploratory analysis on the dataset.	✓
Week 13	Further improve the dataset. Overall analysis & Per Region analysis.	✓
Week 14	Prepare the final presentation Finish the Wikipedia page	✓

Methodology

Data collection

For a start, we scanned a physical French cookbook.

French cookbook.

A page of the scanned book.

The we did a basic OCR for the scanned files. Here is a sample output from OCR.

Sample OCR.

Data digitalization

Template output of the digitization

Data processing

In our project, we will extract the following information from the recipes:

• quantity
• unit
• ingredient

Units 'litre', 'litres', 'l', 'cl', 'dl', 'kg', 'g', 'pincée', 'cuil.', 'cuil. café', 'cuil. soupe', 'cuil. à soupe', 'petite cuil.', 'grande cuil.', 'verre', 'verres', 'petit verre', 'verre à liqueur', 'verres à liqueur', 'tasse', 'tasses', 'bout.', 'bouteille', 'bouteilles', 'grande boîte', 'gousse', 'gousses', 'branche', 'branches', 'membre', 'membres', 'tronçon', 'tronçons', 'tranche', 'tranches', 'tube', 'tubes',

Links

Github repository: Historical Cookbook

Future work

• Build a search engine that would display the recipes and add filters to search them by name, region or ingredients

• User-friendly interface to visualize the results of the analysis

• Comparison with other cookbooks from different periods or different countries