INTRODUCTION
Machine learning is the science and art of programming computers so they can learn from data (Gron, 2019, p. 1). Machine learning is a collection of approaches that automates analytical model building and supports the discovery of insights from data. This assumes that algorithms can learn and improve from data, including examples, direct experience, or instruction; identify relationships; and help make decisions with minimal interventions from the human user. It is the pivotal engine behind artificial intelligence efforts supporting supervised, unsupervised, and reinforced learning algorithms enabling the analysis of massive amounts of data.
While early forms of machine learning have existed for decades, in recent years the applications of machine learning fueled by vast new amounts of big data have grown exponentially. Advertisement placements, dating recommendations, genetic medicines, baseball, stock market investing, and self-driving cars are just a few domains that are heavily driven by machine learning technologies. Ethical issues are emerging from these developments that still have not been fully resolved by society. But one thing appears certain: machine learning is here to stay.
In this course you have read about many of the foundational concepts and techniques of machine learning. This task asks you create a detailed proposal concerning how machine learning techniques could be applied to a relevant data set of your choosing, what possible outcomes might be expected, and why such outcomes could be organizationally significant.
SCENARIO
In this assignment, you will assume the role of a recently hired machine learning engineer in an organization that has been asked to use available data to identify an organizational need that could be solved by machine learning. The organization then needs to outline a plan for designing and executing a machine learning model to solve this organizational need. You will explore available data sets and machine learning algorithms supporting the construction of a model that the organization would use, following the plan that you put together. You will outline the problem, the solution, and the project plan, as well as the framework for evaluating the success of the model and the project.
Project Overview
A. Create a proposal for a machine learning project by doing the following:
1. Describe an organizational need that your project proposes to solve.
2. Describe the context and background for your project.
3. Review three outside works that explore machine learning solutions that apply to the need described in part A1.
Note: These works may include interviews, white papers, research studies, or other types of work by industry professionals. Works that support your research may be identified from various sources, including the WGU Library.
a. Describe how each reviewed work from part A3 relates to the development of your project.
4. Summarize the machine learning solution you plan to use to address the organizational need described in part A1.
5. Describe the benefits of your proposed machine learning
Machine Learning Project Design
B. Describe your proposed machine learning project plan by doing the following:
1. Define the scope of the proposed machine learning project.
2. Explain the goals, objectives, and deliverables for the proposed project.
3. Explain how you will apply a standard methodology (e.g., CRISP-DM, SEMMA) to the implementation of your proposed project.
4. Provide a projected timeline for the proposed project, including the start and end dates for each task.
5. List resources (e.g., hardware, software, work hours, third-party services) and all associated costs needed to implement the proposed solution.
6. Describe the criteria that you will use to evaluate the success of the project once it is completed.
Machine Learning Solution Design C. Describe the proposed machine learning solution you will use to address the organizational need identified in part A1 by doing the following:
1. Identify the hypothesis of the proposed project.
2. Identify the machine learning algorithm(s) (i.e., supervised, unsupervised, or reinforcement learning) you will implement in your proposed solution.
a. Justify the selection of the algorithm in part C2.
