We have a dataset containing 180 email messages. 90 of the email messages were identified as spam and 90 were identified as not spam. Fout features were selected for the task of building a classification model to identify if an email was spam (positive class) or not (negative class). A tabhe explaining the features and some R output is shown below. spaindat = spandat $>.8 mutate(preds = predict( (glm1, type = "response") \%sx round()) im = caret: ; confusionMatrix(data i= factor(spandatSpreds). reference = factor (spandatsis_spam), positive = "1.1) spandat = spandot 10x mutate(preds = predict(glm1, type = "response") os round()) cm = caret:: confusionMatrix(data = factor(spandatspreds), reference = factor ( spomdatsis, 5pam), cmstable positive = "1") Using the confusion matrix output and noting that the positive class is the presence of spam (coded as 1). calculate the following performance me For each, report your answer to 2 decimal places. Given the estimated model output bylow, write down the fitted logistic regression model. summary(glint) at Coll: HA gla(formula = is span - . fanily = binonial, data = spandot) Devtance Residuals: AN Coeffictents: HA (Dispersion parameter for binomial family taken to be 1) AN Nul1 deviance: 249.53 on 179 degrees of freedon * Residuol deviance: 163.36 on 175 degrees of freedom AIC: 173.4 H= In Number of Fisher Scoring iterations: 5 : Edit View Insert Format Tools Table