Machine Learning Question

Description

All the details are in the assessment brief below , it is an easy and very simple project ! provide code and explaination text.

Don't use plagiarized sources. Get Your Custom Assignment on
Machine Learning Question
From as Little as $13/Page

Unformatted Attachment Preview

Reassessment Brief
M505 Intro to AI and Machine Learning
Winter Semester, 2023
Part I: General Information
Module
Term
Assignment Title
Weighting
Distributed on:
To be submitted on:
Submission Method
Length
M505 Intro to AI and Machine Learning
Semester 1, Quarter 1
Individual Final Project
70% Primary Assessment Task
15% Online Assessments
15% Class Participation
W/C November 27, 2023
January 12th, 2024. 18:00 CET
This assignment must be submitted as a Jupyter Notebook (converted
to a *.html file) in the corresponding submission folder to be found
on Canvas.
You must submit your work with an Assessed Submission Form, which
must be completed in full. Please take a screenshot of the signed
Assessed Submission Form and add it to the end of your Jupyter
Notebook as an image. The assignment will not be accepted by the
Registry unless the form is completed correctly.
The submission should contain all the required textual and code
sections in a correct, complete, and concise manner. To avoid
verbosity, please keep the total number of characters in your
submission below 20,000.
Part II: Assessment Details
Primary Assessment You are the newly appointed data scientist of the company. For a given
Task Topic
dataset, you are required to build an end-to-end machine learning
pipeline in a Jupyter Notebook. Your designed and implemented
pipeline will be submitted to the team lead data scientist of the
company.
The notebook should contain the following information:
• A problem statement that elaborates the task. For example,
what is the underlying business problem and why is it
important? How solving this problem will benefit the
company? How would you collect relevant data? How would
you formulate this problem as a machine learning task?
• A data exploration discussion on the characteristics of the
given dataset. For example, how do distributions of the
features and the target class look like? Does this dataset
Assessment
Guidelines
require any sampling or balancing techniques? Which
evaluation metrics are fitting this dataset?
• Data preprocessing and feature engineering steps to prepare
the dataset for learning.
• A model training step to select the best machine learning
algorithm and tune its hyperparameters.
• A model assessment step to evaluate the final performance of
the best trained model.
• A final discussion on the overall pipeline. For example, what are
the overall strengths and limitations of the proposed solution?
What are the implications of the results for the business
problem? What are your data-driven recommendations for
solving the initial business problem? What are the most
informative features of your model? Is your model
explainable?
The submitted Jupyter Notebook should contain both textual sections
and runnable codes in a rational structure. The texts and codes should
be written in a clear and easy-to-follow manner.
All the design decisions should be made in a principled manner. In fact,
all the choices should be justified in the notebook, either by explaining
the intuition or by conducting empirical experiments.
Be creative. Get inspired from any public documents (e.g., blogs,
documentation, open-source projects) but design and implement your
notebook yourself. Reproducing another source will lead to plagiarism
issues.
Please choose a new dataset that was not used in the exercises. Please
clearly mention the URL of your dataset. It allows us later to rerun your
notebook, if necessary, as the dataset is accessible via its URL.
Purpose
The use of generative AI technologies (such as ChatGPT) in your final
assignments is not allowed unless the assessment guidelines explicitly
clarify, under which terms, you are allowed to use these technologies.
Any violation of this rule will result in an investigation of academic
misconduct.
Designing and implementing such end-to-end machine learning
pipelines is one of the key responsibilities of data scientists in practice.
This assignment is designed to assess your ability to build such
pipelines. We are especially interested to see that you can apply
various techniques that you have learned in the module in a systematic
and principled way.
Links to Module
The assignment relates to the following intended learning outcomes
Intended
Learning for the module:
Outcomes
• Critically evaluate the business contexts that can benefit from
machine learning.
• Critically analyse and evaluate core machine learning concepts
and algorithms, including supervised learning and
unsupervised learning.
• Critically design and implement machine learning systems for
various problems using Scikit-Learn.
Special Instructions
Additional
GISMA University rewards in-class participation, and engagement with
Assessment
asynchronous content, at a rate of 30% per module.
Components
Students participating ≥ 80% (factoring on possible extenuating
circumstances) of their synchronous classes as per their due mode of
delivery will gain 15% towards their final module mark.
Students successfully engaging with asynchronous material on the
gamification/microlearning path and completing all summative
assessments in the asynchronous environment will equally gain 15%
towards their final module mark.
Designated asynchronous tasks should be completed by the deadlines
specified by the tutors. Do note that all tasks must be completed by
the deadline applicable for the principal assessment task.
The above also entails that students falling below 80% of participation,
although they will still be allowed to submit, will have their final mark
capped at 85/100. Equally, if they fail to engage with the asynchronous
material and complete the short summative assessments included in
specific checkpoints during each term (usually 4), their module mark,
irrespective of their engagement and participation in synchronous
delivery, will drop by a maximum rate of 15%.
Part III: Marking Criteria / Assessment Criteria
Mark
Weight
Fail
(0 – 49%)
Sufficient
(50 – 59%)
Satisfactory
(60 – 74%)
Good
(75-89%)
Very Good
(90-100%)
100%
5,0
4,0 – 3,7
2,7-3,3
1,7-2,3
1,0-1,3
Marking
Criteria
Does not fulfil
the requirements
Demonstrates
acceptable
knowledge
Demonstrates
substantial knowledge
Demonstrates
comprehensive
knowledge
and
a
and
Demonstrates
comprehensive
knowledge
a
and
of
the
assessment.
understanding of the
subject-matter and
achievement
of
learning
outcomes at low to
average
level of performance.
and understanding of
the subject-matter and
achievement
of
learning
outcomes at average
to
above
average performance
levels.
understanding of the
subject-matter
and
achievement
of
learning
outcomes at well
above average levels
of performance.
understanding of the
subject-matter
and achievement of
learning
outcomes at
high (highest) levels
of performance.
Assessment Criteria Your primary assessment task will be assessed based on the following
criteria:
• The correctness, completeness, and conciseness of runnable codes.
(35%)
• The structure of the report, quality of writing, and critical evaluation
of codes and results in the text. (35%)
Notes
about
Marking
Part IV: Tips for Successfully Engaging with this Assessment
Answer the Question
It may seem obvious, but make sure you are answering the question
you have been set, not the question you would prefer to answer. If the
brief has a number of tasks or parts, answer all of them. Parts that
involve evaluation or analysis are usually longer and worth more marks
than parts that ask for description or explanation. Keep the brief in
front of you and check it regularly.
How
to
Use The assessment criteria document is not usually a guide to the
Assessment Criteria
structure of your assignment. Each section of the criteria is not a
separate paragraph in your assignment, but qualities that you need to
demonstrate throughout. Treat the assessment criteria as a checklist
at the end not as a plan at the beginning. Also, the criteria document
often tells you what to demonstrate (e.g., critical analysis) but not
necessarily how to do it. For how to do it, look back at the skills and
activities you have covered in the rest of the module.
Planning
Preparation
Referencing
Plagiarism
Cheating
Above all, remember this is not a test of how much you know or how
much you have read about the topic. It is a test of how well you can
use your knowledge to answer the specific question set.
and Make sure you attend the lectures, especially the first and the last one,
where we will be ‘unpacking’ this assignment in greater detail.
GISMA Business School requires that students use Harvard
Referencing.
and Your attention is drawn to the University’s stated position on
plagiarism. THE WORK OF OTHERS THAT IS INCLUDED IN THE
ASSIGNMENT MUST BE ATTRIBUTED TO ITS SOURCE (a list of
references and bibliography must be submitted).
Please note that this is intended to be an individual piece of work.
Ensure that you read through your work prior to submission. Action
will be taken where a student is suspected of having cheated or
engaged in any dishonest practice. Students are referred to the
University regulations on plagiarism and other forms of academic
misconduct. Students must not copy or collude with one another or
present any information that they themselves have not generated.
For more information on Plagiarism, please see the relevant section in
your Programme Handbook.

Purchase answer to see full
attachment