I WILL PAY MORE (2* OR MORE) THAN THE BILL THAT THE SYSTEM ARRANGED DUE TO THE AMOUNT OF WORK IT CONTAINS.
Dataset to analyze: https://data.cityofnewyork.us/Public-Safety/Motor-Vehicle-Collisions-Crashes/h9gi-nx95/about_data
Citation style to use (required): ACM -> Link: 
https://www.acm.org/publications/authors/reference-formatting 
I also Attached the resources (Papers I am going to cite)
Review the dataset and decide on a research question that is appropriate to answer by applying the machine learning algorithms. Your research question should be clear, concise and in one sentence.
And IPlease write me a Research Proposal with the research question by 4/7/2024
Your Proposal: an MS Word document using font Calibri 11pts, double spaced to summarize your project idea, dataset to use and algorithms that you will use (approximately 800 – 1500 words). This counts as 10% of your project grade. Your proposal document can be sections I, II, (or maybe III).1 of your Course Project Writeup. (This is due on April 6th, Please hand me this part at that day)
Your Course Project Writeup original document: Structure (Please look at the Screenshot I attached below)
Tips to maximize your project score:
– – –
If the reader thinks multiple times “what are they trying to say here?” your points will be reduced. 
— If the answer to the question “Is this what an entry level data scientist would have done?” is “probably”, you earn
more points 
— If there are no obvious answers to “Is there a clearly better way to perform this analysis, data visualizations
modelling and describe the results?”, and if I learn something new that is attention grabbing, you earn more points. 
One MS Word document using font MS Word Calibri 11pts, double spaced (minimum 2500 words, maximum 4000)
Do the analysis Using Python (Jupyter Notebook)
Review the dataset and decide on a research question that is appropriate to answer by applying the machine
learning algorithms. Your research question should be clear, concise and in one sentence. 
Your Proposal: an MS Word document using font Calibri 11pts, double spaced to summarize your project idea,
dataset to use and algorithms that you will use (approximately 800 – 1500 words). This counts as 10% of your
project grade. Your proposal document can be sections I, II, and III.1 of your Course Project Writeup.  (This is due on April 6th)
Your Course Project Writeup original document: Structure(Please look at the Screenshot I attached below)
Additional to note: When doing machine learning, you should use various methods to ensure the validity of the algorithm you chose (e.g. cross-validation). Not just a Lucky “Result”
Reference & Plagerism
If your submitted work includes content from other sources and it is not properly cited, your submission grade will be
reduced, and your document will be further reviewed by a committee.  
If your submitted work contains anything verbatim from a source, including content generated by LLMs and
Generative AI Tools, it must be enclosed in quotation marks and the proper citation or reference should be included.
If your work includes information from another sources expressed in your own words, the proper citation should be
included but quotes should not be used 
If you write code and part of your code (5 consecutive lines or more) is available publicly online, you must include
the hyperlink to the source in your citation section of your assignment. 
Your work will be processed via Turnitin and other validation tools. A maximum of 10% verbatim from other sources
can be accepted without grade impact, as long as it is in quotes, it is related to concepts (not your dataset and
specific work on it) and the source is properly cited.
Using Jupyter Notebook(s) and Python libraries, create a Flow and write code to answer the
following in this order: 
a. Create a new project named Course Project and load the dataset. 
b. Go through detailed data exploration, statistical analysis, and applicable visualizations to demonstrate that
you can gain a good understanding of the data. 
c. Use judgment and apply appropriate methods to your dataset to address potential challenges related to
missing data, outliers, potential collinearity, etc. and update the dataset in memory 
d. Using the updated dataset, remove 10% of the total observations randomly for validation purposes. 
e. Prepare the remaining of the dataset for training a model using 5 applicable algorithms: subset the data
(break into train and test subsets, transform, scale, etc., as necessary), create and train the respective
models and obtain appropriate metrics. 
f.
Run all models created to answer the research question, reflect on the answers obtained, contrast them
with expected answers based on metrics from the previous step and draw conclusions 
g. Decide which algorithm performs the best and verify its performance, using the 10% of the data you
previously set aside for validation and provide an analysis of what you observe.
h. Draw conclusions and suggested next steps to improve your solution.