Liberty University Statistics K-Nearest Neighbor Classification Essay


Description

k-Nearest Neighbor Classification

The aim of this assignment is to transact k-Nearest Neighbor genus, render the results, and awaken whether or not the instruction generated can be used to address a specific employment tenor.

For this assignment, you achieve use the "Adult Incomes" axioms set from the Topic Materials.

ABC Reconnoitre Company collects axioms via reconnoitres that it then sells to marketing departments. Marketing departments typically do not love detriment axioms. Since reconnoitre takers typically do not love to rejoinder topics in-reference-to their stipend, the one topic usually detriment from the reconnoitre results is, "Is your annual stipend $50,000 or further?"

You are the analyst who has been tasked after a while repartee a way to refer (i.e., fill-in) the rejoinder to the topic, "Is your annual stipend $50,000 or further?" This instruction can best be referd fixed upon how men-folks rejoinder other reconnoitre topics akin to their matrimonial condition, educational flatten, usurpation, and familial interdependence condition. If this weighty topic can be accurately referd, then the compute of the reconnoitre axioms supposing by ABC Reconnoitre Company increases dramatically.

Question 1: Using solely "Marital_Status," "Education," "Occupation," and "Relationship" variables, invent the reckon of neighbors (k) that minimizes the fallacy admonish. Use a place of k betwixt 3 and 10. Include the "k Adnon-interference Fallacy Log" output when submitting the rejoinder.

Question 2: Using the selfselfsame variables and the k selected in Topic 1, rerun the direct neighbor pattern using the characteristic adnon-interference non-interference in the IBM SPSS Modeler. What is the set of variables that minimize the fallacy admonish? Include the "Predictor Adnon-interference Fallacy Log" output when submitting the rejoinder.

Question 3: Using the compute of k and the set of variables that minimizes the fallacy admonish, rerun the k-Nearest Neighbor pattern. What is the genus consultation? Include the pivot consultation output when submitting the rejoinder.

Question 4: Consider the forthcoming specific: Marital_Status=Never-married, Education=Masters, Occupation=Sales, and Relationship=Not-in-family. Fixed on the k-Nearest Neighbor pattern from Topic 3, how would this specific be classified? Provide the predicted proceeds flatten (">50K" or "<=50K") and interpret the rule that you used to state the proceeds flatten. Include the consultation illustrating the axioms when submitting the rejoinder.

Question 5: Describe the pattern fabric rule you used to state whether or not a point reconnoitre taker earned an annual stipend of $50,000 or further. Include discourse of the correctness of the k-Nearest Neighbor pattern and how it can be used in action to refer the rejoinder to the topic, "Is your annual stipend $50,000 or further?"