Project Phase-I Report Outline and Checklist Solution

Description

We strongly encourage you to follow the Phase-I report outline below, as it aligns well with the checklist and grading rubric.

The title / header of your Phase-I report should list (i) the Team-ID of your group¹ and (ii) for each group member the name and Illinois email address.

Report Outline / Checklist

The project instructions (separate document!) describe what exactly you should do as part of

Phase-I of the project. The following outline should be followed for your Phase-I report:

1. Dataset Chosen (5 points)

a. Name the dataset you will be using for your project.

2. Description of Dataset

1. Here you will provide an ER diagram, an ontology, or a detailed database schema (10 points), and

2. a narrative description of the dataset covering structure and content (15 points)

3. Use Cases

1. “Zero cleaning” use case U0: data cleaning is not necessary (5 points)

2. “Main” use case U1: data cleaning is necessary and sufficient (20 points)

3. “Never enough” use case U2 : data cleaning is not sufficient (5 points)

4. Data Quality Problems

1. List obvious data quality problems with evidence (examples and/or screenshots) (20 points)

2. Explain why / how data cleaning is necessary to support the main use case U1 (10 points)

5. Initial Plan for Phase-II (10 points)

1. Below is a possible plan, listing typical data cleaning workflow steps. In your Plan for Phase-II, fill in additional details for the project steps as needed. In particular, include who of your team members will be responsible for which steps, and list the timeline that you are setting yourselves!

   • S1: Review (and update if necessary) your use case description and dataset description

   • S2: Profile D to identify DQ problems: How do you plan to do it? What tools are you going to use?

   • S3: Perform DC “proper”: How are you going to do it? What tools do you plan to use? Who does what?

   • S4: Data quality checking: is D’ really “cleaner” than D?

1. • • Develop test examples / demos

1. • S5: Document and quantify change (e.g. columns and cells changed, IC

violations detected: before vs after, etc.)

• In addition you can give yourself a (cute) team name for ease of identification 😉