Welcome to my personal page

Admittedly only had 54% accuracy but was my first project and introduction to Data Science. Where I also was pushed to learn R. Encode categorical data into numeric data (one-hot encoding) and filling in some missing data using some common sense and a little analysis about the passengers and that time in history.

Let us be honest not everyone can get a loan or the amount they want approved and if you happen to work at a bank then this model is for you. The whole process I took was looking at the statistical summary, distribution of the variables, encoded categorical variables, filled in in missing data via pivot table, treated outliers, and finally modelling the data.

Not all grocery stores will generate the same amount of sales every month. Factors like location and if it is a franchise matter. My thought process here was hypothesis generation, data exploration, data cleaning and finally modelling. Used a linear/ridge regression model and a decision tree/random forest.

I can be quite lazy at times. So, when I know there are obvious interview questions that are asked most of the time. I created a chatbot to do the talking for me...and to possible convince you, the reader, to hire me. Developed using a Deep Neural Network so that you do not even have to type the entire question just the keyword/topic about eh question. For example, typing in "yourself" will automatically answer the question tell me about yourself.

Used yolov4 as a pre-trained real time object detection model. I plan to further train it on detecting weapons and license plates. Why? So that if any of those are detected by a security camera the proper authorities will be immediately notified. I also eventually want to train it to detect sign language and expand it to create an aimbot by detecting weapons as stated previously.

One of my goals in life is to own a house. So, I would like to know what factors go into determining a house price. The two most important factors are the number of rooms (positive correlation) and location in terms of number of "lower class" people. Take that how you will.

What is the best way to reach a customer? That was the question I tried to answer. Turns out customers with the highest lifetime value do not need to be reached and will reach out to you instead.