Companies have a tendency to ask for how much an applicant is comfortable with as salary. This can be decided based on a number of factors plus other considerations. In predicting this, it allows applicants to estimate how much a company's worth and if they can afford to work there given the company's expectations and environment. In this project, we look at some of the features considered and how they affect whether someone earns less than 50k within their respective companies.

Adult income dataset https://www.kaggle.com/datasets/wenruliu/adult-income-dataset

In the original dataset, there werer 48,842 rows and 15 columns.

To prepare this data, the data was cleaned, and the following processes were performed:

• To visualize the data for explantory purposes, bargraphs were used.
• The bargraph was chosen to show how the working class categories varry with hours per week worked based on whether an income of above 50k or not is earned.
• The bargraph shows that in all the working classes, those who work for more hours per week earn above 50k compared to those who don't with the exception of those who've never worked it seems that they all receive below 50k. That might be the default amount or something.

This bargraph shows that the older someone is irrespective of their education level, they earn above 50k. Its most likely an advantage of being more experienced or holding more responsibilities.

Metrics for fine tunned Logistics regression model with Principel Component Analysis applied

Training data

Test data

From the stakeholder's perspective, false positive would mean that individuals might end up spending more than they need to like in projects that end up running bankrupt while a false negative would mean that individuals might end up restraining themselves on a tight budget yet they could afford to be more comfortable with their expenditure.

As seen from the support columns of each metric, the dataset much less number of people with incomes above 50k, so the model had to be tunned to reemedy that.

• I'd advise the stake holders to keep a keen eye on their incomes and avoid assumptions, it would be alarming to make more losses than initially anticipated.
• To secure their place in the greater than 50k income category, i'd advise the stakeholders to keep on upgrading their features for such things are also considered.