Machine Learning Project

To bring together and apply the various topics covered in this course, you will work on a machine learning project.

• Implementation of ML Project: You will implement your ML solution using 2 APIs:

  • SciKit
  • Spark ML

• The goal of the project is to go through the complete knowledge discovery process to answer one or more questions you have about a topic of your own choosing.

• You will acquire the training data, cleaning data, query data, formulate a question (or questions) of interest, perform the building model and data analysis (using machine learning algorithms), and

• Communicate the results to the class by a series of writings and presentations

Project Team

Students are encouraged to work in teams of 3–4 (with a maximum of 4 members) for the course project.

Project Milestones

• There are a few milestones for your final project.

• It is critical to note that no extensions will be given for any of the project due dates for any reason.

• Projects submitted after the final due date will not be graded.

• If you anticipate any issues you need to send an email to the instructor at least one week in advance.

Due Dates

Unless otherwise announced, all submission deadlines are at 11:59pm PST on the assigned due date.

Date	Deliverable
January 23, 2020	Project proposals presentations
January 28, 2020	Project proposals due
Skipped, TBDL	Draft paper due
In progress	Project presentations
March 12, 2020	Final deliverables due

Deliverables

There are several deliverables for your project that will be graded individually to make up your final project score.

Proposal

You start your project by forming your groups and letting the instructors know what topic you are interested in exploring. The proposal should be an informal written document submitted to the instructors in person or via email. The instructors will schedule a project review meeting with you during the week marked in the schedule. Make sure all of your team members are present at the meeting.

Project Presentation

Each group will explain their project in a 20 minutes presentation to the class. Presentations should clearly convey the project ideas, methods, and results, including the question(s) being addressed, the motivation of the analyses being employed, and relevant evaluations, contributions, and discussion questions.

Project Paper

The projects will be concluded with project paper. Your paper should summarize your steps in developing your solution, including how you collected the data, the method you used, and the insights obtained. The paper should be submitted in Word or PDF format. Optionally, you may submit a draft paper (in person or via email) by the listed deadline to receive feedback from the instructor prior to the final paper deadline.

Grading

• 70% of the project grade will be based on your project paper (data selection, cleaning, ML algorithms, results, accuracy)
• 20% of the grade will be based on your project presentation.
• 10% of the grade will be based on your final exam.

The project proposal will be graded as follows:

• Title of Project: 5% What's the title of the project?
• Project Plan: 30% What do you plan to do?
• Data Sources and ML algorithms: 20%
  • What data do you plan to use?
  • From where will this data come?
  • Selection of ML algorithms
• Proposed Evaluation: 30%
  • How do you plan to evaluate your proposed ML model?
  • How will you determine whether the model/method is successful?
• Writing Quality: 15%
  • Clarity of expression (5%)
  • Organization (5%)
  • Grammar (5%)

The project presentation will be graded as follows:

• Introduction: 15%
  • Provide context.
  • What questions are being addressed?
• Solution/Method: 30%
  • What did you do?
  • Why did you choose this method/model?
  • What tools and techniques did you use?
• Data and Experiments: 10%
  • What data did you use?
  • Are your experimental models/methods reliable?
• Evaluation and Results: 30%
  • What evaluation did you do?
  • Do your conclusions match your results?
• Presentation Quality: 15%
  • Clarity of speaking (5%)
  • Organization (5%)
  • Visuals (5%)

The project paper will be graded as follows:

• Introduction: 15%
  • Provide context and motivation.
  • What questions are being addressed?
  • Why are these questions interesting or important?
  • What is the story?
• Related Work: 10%
  • What other methods have addressed these or similar questions?
  • How do these methods differ from your method?
• Solution/Method/model: 25%
  • What did you do?
  • What tools and techniques did you use?
  • Was any innovation attempted?
  • What models did you try?
  • How did you improve the model?
• Data and Experiments: 15%
  • What data did you use?
  • Are your experimental methods reliable?
  • What preprocessing was done the data?
  • Did you drop any columns? why?
  • Did you add any derived columns? why?
  • Data cleaning and normalization? how?
• Evaluation and Results: 20%
  • Did you properly evaluate your experiments/models?
  • Did you test for statistical significance?
  • Do your conclusions match your results?
  • Can your solution be used in real-life?
• Writing Quality: 15%
  • Clarity of writing (5%)
  • Organization (5%)
  • Grammar (5%)

Submission Instructions

To submit your project deliverables,

• create a GitHub repository and place your deliverables in designated folders (place your files in specific folders).

• The folder should include

  • Your project paper (Word or PDF format),
  • Presentation materials, and
  • All data, code and output used to generate results.

• If I cannot access your work because these directions are not followed correctly, I will not grade your work.