Coursework Assessment

Coursework AssessmentOne of the main objectives of this module is to help you gain hands-on experience incommunicating insightful and impactful findings to stakeholders. In this coursework, you willuse the tools and techniques you learned throughout this module to train few machine learningmodels on a dataset that you feel passionate about, select the techniques that best suits yourneeds, and communicate insights you found from your modeling exercise.After going through some guided steps, you will have insights that either explain or predict youroutcome variable. As a main deliverable, you will submit a report that helps you focus onhighlighting your analytical skills and thought process.You are expected to leverage a wide variety of tools such as Jupyter notebook, Python and therelevant machine learning libraries (Keras, Tensorflow, Pytorch, etc.), but your report shouldfocus on present findings, insights and next steps. Before you begin, you will need to choose adata set that you feel passionate about. This can be a data set similar to the data you haveavailable at work or data you have always wanted to analyse. For some people this will besports data sets, while some other folks prefer to focus on data from a datathon or data forgood. Data for Good, inspired by, brings together leading data scientists with high impactsocial organizations through a comprehensive, collaborative approach that leads to shared insights,greater understanding, and positive action through “data in the service of humanity”. Below are thelinks to 5 data sets:1. Fortune 500. URL: AT&T stock price data. URL: COVID-19 variants. URL: Leukemia gene expression. URL: Stock exchange data. URL: you have selected a data set, you will produce the deliverables listed below:A. Main objective of the analysis that specifies whether your model will be focused onprediction or interpretation.B. Brief description of the data set you chose and a summary of its attributes.C. Brief summary of data exploration and actions taken for data cleaning and featureengineering.D. Summary of training four machine learning models. For regression, the model will bemultiple linear regression, polynomial regression, LASSO regression and ridgeregression. For classification, the model will be multilayer perceptron, convolutionneural network and variants of multilayer perceptron and convolution neural networksuch as ResMLP and GoogLeNet. For clustering, the model will be K-means, hierarchicalclustering, DBSCAN and OPTICS.E. A paragraph explaining which of your models you recommend as a final model that bestfits your needs in terms of accuracy and explainability.F. Summary Key Findings and Insights, which walks your reader through the main driversof your model and insights from your data derived from your models.G. Suggestions for next steps in analysing this data, which may include suggesting revisitingthis model adding specific data features to achieve a better explanation, a betterprediction, etc.Compulsory: Please submit a PDF file containing the deliverables A to G. You should includethe visuals from your code output, but this report is intended as a summary of your findings,not as a code review.OptionalYou may submit your code as a python notebook (.ipynb file) or as a print out in the appendix ofyour main PDF report.GradingThe grading will be based on 5 main points:1. Does the report include a paragraph detailing the main objective(s) of this analysis [10marks]?a. This report is missing a planning section for the data analysis. [0 marks]b. Yes. This plan includes a detailed subtask section or a good vision of what ispossible to do with this data set. [5 marks]c. This plan exceeds expectations. In addition to plan out subtasks and vision forthis analysis, it also anticipates possible snags that might be incorporated intopreliminary hypothesis of the data. [10 marks]2. Does the report include a section describing the data [10 marks]?a. There is no summary or it is hard to put together what variables are available orhow they might be used. [0 marks]b. There is a basic summary, like a data dictionary. [5 marks]c. The summary of the data is presented with graphs of distributions and plots thatshow the relation between features and the outcome variable. [10 marks]3. Does the report include a section with variations of machine learning models andspecifies which one is the model that best suits the main objective(s) of this analysis [10marks]?a. No. It is not clear if a machine learning model was used in this analysis, or themachine learning model is missing. [0 marks]b. Yes. Four machine learning models are included and it discusses findings andresults appropriately. [5 marks]c. Yes. There are four machine learning models. One of them is presented as thebetter alternative and some findings are presented. The findings should includewhat variations of a machine learning model should be considered (testing splits,cross validation, polynomial features, regularized regressions, cluster selection,etc.). [10 marks]4. Does the report include a clear and well presented section with key findings related tothe main objective(s) of the analysis [10 marks]?a. No. There are no takeaways, insights or findings about this problem. [0 marks]b. Yes. Some takeaways and findings derived from the model are presented. [5marks]c. Yes. Takeaways and findings derived from the model are well presented. [10marks]5. Does the report highlight possible flaws in the model and a plan of action to revisit thisanalysis with additional data or different predictive modeling techniques [10 marks]?a. No. There is no mention of possible flaws or plans to revisit the analysis. [0marks]b. Yes. There is some discussion presented on possible flaws of this model and aplan to revisit this with additional data or different predictive modelingtechniques. [5 marks]c. Yes. There is a comprehensive list of possible flaws of this model and a detailedplan to revisit this with additional data or different predictive modelingtechniques. The quality of this section gives it full marks. [10 marks]FAQsQ1: Do I have to come up with my own data set?Ans: You are highly encouraged to find a data set you feel really passionate about. This willhelp you showcase analytical work that truly matches your skills. But if you prefer, you can usesome of the data sets from this module.Q2: Is it OK to choose the same data set as someone else?Ans: Yes, more than one person can analyse the same data set. Most likely your insights will bedifferent from your peers and you will still be able to showcase your own talent as a uniquesolution.Q3: Do I have to train more than four machine learning models?Ans: You are required to train four machine learning models to highlight which modelimproved your prediction or interpretation.Q4: Is this an individual assignment?Ans: This is an individual assignment. You can ask for help or assistance on technical issues andgeneral direction of your analysis, but the interpretation of the analytical output and thewriting of the report should be your own.

YOU MAY ALSO READ ...  Model School Technology Programs-An Emerging Technology in China

Assignment status: Already Solved By Our Experts

(USA, AUS, UK & CA Ph. D. Writers)


Order from Australian Expert Writers
Best Australian Academic Writers