MachineLearning代写：BUAN6341MachineLearningProblem

发布日期: 2023-04-17

代写机器学习作业，自选数据以及算法，完成整个项目。

COURSE PROJECT GUIDELINES

This course project offers an opportunity for you to explore an interesting
and practical machine learning problem in the context of a real-world dataset.
A typical project consists of picking an interesting dataset, applying one or
more appropriate and well-known machine learning techniques for an interesting
task as baselines, and extending these baselines in creative and interesting
ways.
Projects can be done in teams of at most 4 students. Team members are
responsible for dividing up the work equally and making sure that each member
contributes. The course project will be worth 20% of your final grade. In this
document, we describe the detailed requirements in completing this course
project, as well as some suggestions of choosing ML-related datasets and
questions.

DELIVERABLES

This course project requires 3 delivarables:

Proposal
Class Presentation
Final Report

PROPOSAL

A one-page proposal. You have to submit your proposal via eLearning. Only one
copy per group is required. You are encouraged to discuss your ideas with the
instructor or TA before submitting the proposal.
A complete proposal should include the following information:

Project title
Project descriptions (e.g., why it is an interesting question in the real world)
Dataset(including both sources and brief description, such as data fields and summary statistics)
Teammates and work division
Potential methodologies (i.e., discuss the ML algorithms you plan to use, including at least one supervised and at least one unsupervised learning techniques)

CLASS PRESENTATION

All students should present their project in class. Each project has 20-min
presentation time plus approximately 5-min Q&A time.
All students should be present in the two presentation classes. Each
presentation will be evaluated by all other teams.

Slides - The quality of the slides. Do they get the points across? Are they appealing and attention getting? Do they facilitate an effective presentation?
Presentation - The quality of the delivered presentation. How were the points delivered? Was the communication persuasive and informative? Did the tone grab and keep the audience’s attention? Did the content flow well? Were the speakers effective in delivering the content?
Content - Was the question they aimed to solve interesting? Were the methodologies they chose proper? Were the analyses process accurate and correct? Did the content reflect an understanding of the topic and a high degree of analysis?
Questions - How did the group handle questions? Were they prepared with good answers? Did they handle the questioners with respect and with an eye to convincing the audience of their views?
Extra - Did the presenters do something “extra” that caught the attention of the audience or kept the audience entertained and interested?

FINAL REPORT

The final report is without page limits. A complete final report should
include:

A detailed description of the entire project (e.g., project description, dataset description, methodologies, results, and implications)
Your python codes (with comments)

GRADING

Your grade will be determined by four components: (1) your proposal (counts
approximately 10%); (2) peer evaluations (within and between groups); (3) your
class presentations; and (4) your final report.
Some general tips:

higher grades will be given to projects that analyze bigger datasets compared to small ones
higher grades will be given to projects that explore multiple up-to-date machine learning models
higher grades will be given to individuals who provide critical insights to other teams (members) during the Q&A process

SUGGESTIONS

Note: This is a challenging and also open-ended project, with no pre-defined
“correct” answer. It is up to you to locate a dataset (or any new trend) that
is interesting and possible to analyze in a meaningful manner. Projects that
combine multiple datasets and multiple ML-related methodologies will receive
higher grade.
Here are links to several interesting datasets: