辅导 program、讲解 Python/Java程序
Assignment 2: Predict Students Dropout and
Academic Success
Assigned: 28/11/2024 Due: 22/12/2024
Overview
Welcome to the Student Performance and Retention Prediction Competition! In this
challenge, you will work with a real-world dataset to predict student outcomes, such as
whether a student will drop out, remain enrolled, or graduate. The competition is designed
to allow you to practice Exploratory Data Analysis (EDA), visualization, and machine
learning to gain insights and build predictive models.
This dataset comes from a higher education institution and contains multiple features
related to:
• Demographic Information: Age, gender, socio-economic background.
• Academic Performance: Grades and results from the first two semesters.
• Behavioral and Social Data: Attendance records, participation in extracurricular
activities.
You will use this dataset to predict one of the following outcomes for each student:
• Dropout: The student has withdrawn from the program.
• Enrolled: The student remains enrolled in the program but has not graduated.
• Graduate: The student has successfully completed the program.
Competition Entrance
https://www.kaggle.com/t/728f638f02854aab8f16be47cbeaef6b
Tasks 1 (60 Marks)
1. Create an account at https://www.kaggle.com/. You MUST set a TEAM NAME with
the format in (Student ID). Example: 2136673.
2. Create a notebook on Kaggle. Conduct Exploratory Data Analysis and data processing,
train and validate your models, and generate your ‘submission.csv’ on test data using
the notebook on Kaggle. Add necessary comments to your notebook. Download the
notebook from Kaggle and submit it to Learning Mall, with the name in
(Name_Student ID). (30 Marks)
3. Submit your predictions (‘submission.csv’) for the test solution to Kaggle. Also, you are
required to include your Kaggle score in your report (see below in Task 2). (30 Marks)
Tasks 2 (40 Marks)
Write a 1-page report, which must contain 2 or 3 tables or figures.
Name your report with Name_Student ID.
Submit your report to Learning Mall.
The report must cover:
• Introduction: (4 Marks)
What is the background of this project? How is it related to Big Data?
• Methodology: (8 Marks)
A. Data Preprocessing
What are the steps of data pre-preprocessing explored before training? Data
visualization, data cleaning and reduction, normalization and discretization,
feature selection, imbalanced data, etc. No need to cover all of them.
B. Classification Algorithm
How does it work? Explain the algorithm or framework.
• Results: (10 Marks)
Are there other competitor models for this project? How does it compare to your
technology?
• Discussion: (4 Marks)
What are the good aspects, and what are the bad aspects? Be sure to add a
sentence on “contributor thoughts:” What are your own unique thoughts on the
pros and cons of the technology? Do you envision an extension that might be
helpful?
• Conclusion: (4 Marks)
Summarize the 2 to 4 points you think are most important.
Concise, information-rich content. For each of the sections above, you will not simply
be graded on having content but on the quality of the content and how well it answers
the questions in concise, clear, and engaging terms.
Style. (10 Marks)
In order to make your report consistent and visually appealing, as well as to make the
evaluation of your work fairly, each page should be conformed to the following
specifications:
• Margins: approx. 0.5” on all 4 sides.
• Columns: 2 with approx. 0.3in margin; justified text
• Fonts:
• Body text: Times New Roman, 11pt.
• Section headings: Calibri 13pt bold-Italic
• Within captions, tables, figures, or images: Calibri 9-11pt.
• Line Spacing:
• Body text: Single (1.0)
• Section headings: 6pt spacing above heading
Academic Honesty. Copying chunks of code or problem-solving answers from other
students, online or other resources is prohibited. You are responsible for both (1) not
copying others’ work, and (2) making sure your work is not accessible to others.
Assignments will be extensively checked for copying of others’ work. Problem-solving
solutions are expected to be original, using concepts discussed in the book, class, or
supplemental materials but not using any direct code or answers. Please see the syllabus
for additional policies.