Explain what you did and how they worked.

Learning Goal: I’m working on a machine learning exercise and need support to help me learn.ObjectiveThe goal of this homework is to give you exposure to the practice of training and testing machine-learning models. In particular, you need to build a logistic regression classifier. The performance of those classifiers should be measured by accuracy. The task is to classify patients into high medical spending and low medical spending. The primary goal of the assignment is for you to learn to experiment with feature engineering, and to review some concepts we’ve covered in class.DataThe data (a1_data.csv) for this assignment was prepared for you. The data preprocessing steps explained in the linear algorithms notebook (week 4), including combining CHBMIX42 and BMINDX53, removing outliers, and creating dummy variables, were applied. You should use this file as a data source. What to doThis is a fairly open-ended assignment, but I would recommend applying three different feature engineering methods. The first two of them are introduced in the section about common mistakes in linear regression.LinearityAlthough linear regression and logistic regression assume a linear relationship between dependent and independent variables, the actual data distribution might not follow linearity. If this is the case, you may apply polynomial terms. Please see p.10 in the week 4 slides.
MulticollinearityAlthough linear regression and logistic regression assume no or little multicollinearity between independent variables, it might not be true in your data. If you find multicollinearity between independent variables, then you can multiply them. Please see p.11 to p.13 in the week 4 slides.
Noisy featureIn most cases, having more independent variables (i.e., features) helps to build a better model. However, if there is only a random pattern between an independent variable and a dependent variable, then it might be better to drop the independent variable.
By mixing these three strategies, try to beat the baseline classifier with an accuracy of 0.63 and show your work.What to submitIt might be hard to improve the performance of logistic regression classifiers. Therefore, it is important to explain what you did and how they worked. Taking into consideration everything you tried and whether or not it worked, provide a discussion of your overall results. Did you notice any method that particularly worked or not? Do you have any ideas for why these happened? What did you learn? To demonstrate your work, you need to write a report (a .docx file). Please include your best accuracy for A1 in your report and what method you applied.Please attach a .ipynb file where you include the feature engineering enabled your best classifier. Please note that you should not include all methods you tried. Only the best method and coding for that need to be submitted in the notebook. Please use the a1.ipynb file distributed together.In short, you need submit a report (.docx) as well as a notebook (.ipynb).Grading CriteriaThe performance will be counted as 40% of the grade for assignment 1. Your efforts you made, logic to build the best classifier, and reflection on methods you applied will be counted as 60% of the grade. RubricReflection PaperReflection PaperCriteriaRatingsPtsThis criterion is linked to a Learning OutcomeContentWhether to summarize, synthesize, and apply critical thinking50 ptsFull MarksInterprets topic in accurate and insightful ways. Uses information thoughtfully, in a way that are factually relevant and accurate. Composition demonstrates analysis and might offer alternative thoughts or creative viewpoints based on concrete evidence.37.5 ptsProficientAccurately interprets topic; uses main points of information from assigned texts. May repeat the ideas of theorists but attempts to offer new insight; but does not provoke significant new thinking or further consideration.25 ptsNeeds WorkMakes errors in interpreting topic. Opinion based and superficial commentary only.50 ptsThis criterion is linked to a Learning OutcomeIntegration of external resources/referencesProper in-text citations from readings, and textbook30 ptsFull MarksExternal resources/references are used to support the context very well.15 ptsProficientSome external resources/references are used to support the context, but more references should be used.0 ptsNo MarksNo external resources/references are used to support the context.30 ptsThis criterion is linked to a Learning OutcomeStructureWriting, Grammar, APA style20 ptsFull MarksThe following writing instructions are kept very well: 400-600 words, (12-point font, double-spaced), APA 7th edition style, 4 to 5 paragraphs, introduction, two or three paragraphs about what you learn, and conclusion.10.67 ptsProficientSome following writing instructions are kept kept but there are still some mistakes: 400-600 words, (12-point font, double-spaced), APA 7th edition style, 4 to 5 paragraphs, introduction, two or three paragraphs about what you learn, and conclusion.0 ptsNo MarksThe following writing instructions are not kept: 400-600 words, (12-point font, double-spaced), APA 7th edition style, 4 to 5 paragraphs, introduction, two or three paragraphs about what you learn, and conclusion.20 ptsTotal Points: 100
Requirements: not specified