Comprehensive Foundations of Data Science Project Assignment

Objective: Utilize a selected dataset to conduct a thorough Foundations of Data Science project, covering data exploration, cleaning, visualization, basic inferential statistics, and predictive I

analytics concepts.

Total Marks: $25$

Project Tasks:

$1.$ Dataset Selection and Business Problem Definition $(1$ Mark$)$

Select a dataset and define a clear business problem or question that your analysis

aims to address.

$2.$ Data Overview $(1$ Mark$)$

Conduct a preliminary examination of the dataset. Describe its size, nature of variables $($quantitative vs$.$ qualitative$),$ and the potential value it brings to solving the business problem.

$3.$ Data Types Identification $(1$ Mark$)$

Classify the dataset's variables into their respective data types $($numerical$,$ categorical, ordinal, etc.$).$

$4.$ Missing Values Assessment $(1$ Mark$)$

$$ Identify any missing values in the dataset. Provide a summary of the missingness pattern observed.

$5.$ Data Cleaning Strategy $(1$ Mark$)$

Propose a strategy for handling missing data $($deletion$,$ imputation, etc.$)$ and justify your approach.

$6.$ Outliers Detection $(1$ Mark$)$

Employ a method to detect outliers in the dataset. Briefly describe how you identified them.

$7.$ Handling Outliers $(1$ Mark$)$

Decide on a strategy for dealing with the detected outliers. Explain your choice and its implications for the analysis.

$8.$ Data Transformation $(1$ Mark$)$

Perform necessary data transformations $($normalization$,$ scaling, etc.$)$ to prepare the data for analysis. Justify why these transformations are required.

$9.$ Categorical Data Encoding $(1$ Mark$)$

Encode categorical variables as needed for analysis. Explain your encoding choices.

$10.$ Exploratory Data Analysis $($EDA$)$ Introduction $(1$ Mark$)$

Begin EDA by summarizing key statistics $($mean$,$ median, mode, standard deviation$)$ for at least three variables.

$11.$ Visual EDA: Distributions $(1$ Mark$)$

Create visualizations $($histograms$,$ box plots$)$ to explore the distribution of key variables.

$12.$ Visual EDA: Relationships $(1$ Mark$)$

Generate scatter plots or correlation heatmaps to investigate relationships between variables.

$13.$ Identify and Interpret Key Relationships $(1$ Mark$)$

Based on the visualizations, identify two key relationships in the data. Discuss their potential impact on the business problem.

$14.$ Descriptive Analytics Summary $(1$ Mark$)$

Provide a concise summary of the descriptive analytics findings, highlighting insights relevant to the business problem.

$15.$ Formulate Statistical Questions $(1$ Mark$)$

Based on your EDA, formulate two statistical questions that could further inform the business problem.

$16.$ Hypothesis Testing Plan $(1$ Mark$)$

For one statistical question, outline a plan for a hypothesis test. Specify the null and alternative hypotheses, significance level, and the test statistic to be used.

$17.$ Inferential Statistics Conceptual Application $(1$ Mark$)$

Conceptually apply inferential statistics to estimate a parameter or make predictions about the population based on your sample. No calculations required; explain your thought process.

$18.$ Introduction to Predictive Modeling $(1$ Mark$)$

Identify a potential outcome variable for predictive modeling. Justify its selection based on your business problem and EDA findings.

$19.$ Feature Selection Rationale $(1$ Mark$)$

$.$

Discuss how you would select features for the predictive model. Consider correlations, importance, and relevance to the outcome variable.

$20.$ Feature selection process $(1$ Mark$)$

Select the features using at least $3$ different model driven techniques

$21.$ Predictive Model Choice $(1$ Mark$)$

Choose a predictive modeling technique suitable for your data and business problem $($e$.$g$.,$ linear regression, logistic regression$).$ Explain why this model is appropriate.

$22.$ Data Visualization for Insights $(1$ Mark$)$

Create a dashboard layout or a set of visualizations that could provide actionable insights to a business stakeholder. Describe how these insights address the business problem.

$23.$ Presentation of Findings $(1$ Mark$)$

Summarize the key findings from your project, focusing on how they address the business problem. Include any recommendations or potential actions.

$24.$ Interpretation of Predictive Model Output $(1$ Mark$)$

Once you've conceptually chosen a predictive model, describe how you would interpret the model's output to derive actionable business insights.

$25.$ Conclusion $(1$ Mark$)$

Conclude your project by summarizing the key insights and their implications for the business problem. Reflect on the limitations of your analysis and potential improvements.