Data Integration Project

This repository serves as template for the students project accompanying the "Data Integration" lecture at University of Marburg.

Students can fork or copy this repository to kick-start their project.

Structure

This data integration project has four components: preparation, data integration, data cleaning, and showcase.

0_datasets: We recommend to read and write source, intermediate, and final datasets from 0_datasets directory. Its content is gitignored (i.e., it will not be tracked by Git), use Hessenbox to sync-and-share your datasets.
1_preparation: Code to extract and load your datasets into a common space (e.g., sqlite or Jupyter notebook). You might also want to write code to explore the datasets.
2_integration: Experiments to integrate your datasets and a program that implements your final integration pipeline.
3_cleaning: Experiments to clean your integrated dataset and a program that implements your final data cleaning pipeline.
3_showcase: Code to implement your showcase. You can already run it against the results of your data integration pipeline, and, finally, against your integrated and cleaned data.

All code documentation and instructions should be placed in this README.md; feel free to erase this intro text.