Convert demog_genf and demog_genm into a single categorical variable and name it as 'gender'. Then, drop demog_genf and demog_genm columns from the dataset. Ensure that both the 'gender' and demog_ho variables are set to the categorical data type. Generate Python code to create scatter plots for each numerical feature versus the target variable int_tgt, and examine the correlation matrix of these features. When deciding on the most suitable features for model1, solely base your selection on the insights gained from the scatter plots and the correlation matrix. Specifically, identify three numerical features that demonstrate a clear linear relationship with int_tgt through the scatter plots. Additionally, ensure these three chosen features exhibit low inter-correlation by referring to the correlation matrix. Aim for correlation coefficients less than 0.5 between the selected features to minimize multicollinearity in your linear regression model. After completing your analysis, summarize your reasoning for selecting these particular features for model1 in a paragraph. Focus on the evidence of linearity observed in the scatter plots and the low correlation scores among the chosen features, explaining why they are deemed optimal for modeling int_tgt. Document your explanation under 'Task 1' in the provided Word document.