alecngai

01-Kickstarter-analysis

analyzing kickstarter fundraising data (Kickstarter_Challenge.xlsx) to provide clarity on two main issues, outcomes based on launch date, and Outcomes based on Goals.

02-Stock_Analysis

03-Election_Analysis

04-School_District_Analysis

05_Pybar_Analysis

07_Pwelett_Hackard_Analysis

We will complete two technical analysis using SQL and PGAdmin4, using the data proivded to us we will achieve two results: The Number of Retiring Employees by Title The Employees Eligible for the Mentorship Program

08_Movies_ETL

tasked with creating datasets for a hackingathon, the client "Amazing Prime" wants a dataset that is always up to date and is automated in such a way that we extract, transform and load the dataset into a clean workable dataset. We do this in four tasks: write an ETL function to read three data files, extract and transform the Wikipedia data, extract and transform the Kaggle and rating data, load the data to a PostgreSQL Movie Database.

14_Bikesharing

Using Tableau we are able to analyze the cities bike sharing platform in many aspects.

17_Credit_Risk_Analysis

Credit risk is very tough to predict. In this project we want to take a look at how all the factors in our loan_stats csv help predict whether someone is low or high risk status. One method that data scientists use for this type of issue is creating a model and then evaluate and train the models that they create. In this specific project we are using imbalanced-learn and scikit-learn libraries to build models and evalute them using a resampling method. In the first couple of models I oversampled the data using randomoversampler and smote algorithms and undersample the data with the clustercentroid algorithm. In the remaining models I used a combination approach to over and undersample the data using smoteenn. Finally, I compared two machine learning models that minimize bias, balancedrandomforestclassifier and easyensembleclassifier.

Predicting_Canadian_Job_Vacancies

Understand the Canadian labor market and predict how the job market would look like in the near future. It is important to know how the labour market have changed over the past years and how it will look like in the upcoimg years. Understanding the labour market and looking at potential vacancies could go a long way in avoiding situations such as economic collapse and could also potentially assist with education planning.

alecngai/Predicting_Canadian_Job_Vacancies

Understand the Canadian labor market and predict how the job market would look like in the near future. It is important to know how the labour market have changed over the past years and how it will look like in the upcoimg years. Understanding the labour market and looking at potential vacancies could go a long way in avoiding situations such as economic collapse and could also potentially assist with education planning.

alecngai/01-Kickstarter-analysis

analyzing kickstarter fundraising data (Kickstarter_Challenge.xlsx) to provide clarity on two main issues, outcomes based on launch date, and Outcomes based on Goals.

alecngai/02-Stock_Analysis

alecngai/03-Election_Analysis

alecngai/04-School_District_Analysis

alecngai/05_Pybar_Analysis

alecngai/07_Pwelett_Hackard_Analysis

We will complete two technical analysis using SQL and PGAdmin4, using the data proivded to us we will achieve two results: The Number of Retiring Employees by Title The Employees Eligible for the Mentorship Program

alecngai/08_Movies_ETL

tasked with creating datasets for a hackingathon, the client "Amazing Prime" wants a dataset that is always up to date and is automated in such a way that we extract, transform and load the dataset into a clean workable dataset. We do this in four tasks: write an ETL function to read three data files, extract and transform the Wikipedia data, extract and transform the Kaggle and rating data, load the data to a PostgreSQL Movie Database.

alecngai/14_Bikesharing

Using Tableau we are able to analyze the cities bike sharing platform in many aspects.

alecngai/17_Credit_Risk_Analysis

Credit risk is very tough to predict. In this project we want to take a look at how all the factors in our loan_stats csv help predict whether someone is low or high risk status. One method that data scientists use for this type of issue is creating a model and then evaluate and train the models that they create. In this specific project we are using imbalanced-learn and scikit-learn libraries to build models and evalute them using a resampling method. In the first couple of models I oversampled the data using randomoversampler and smote algorithms and undersample the data with the clustercentroid algorithm. In the remaining models I used a combination approach to over and undersample the data using smoteenn. Finally, I compared two machine learning models that minimize bias, balancedrandomforestclassifier and easyensembleclassifier.

alecngai/06_World_Weather_Analysis

alecngai/09_Surfs_up

alecngai/10_Mission_to_Mars

alecngai/11_UFOs

alecngai/12_Plotly_Charts

alecngai/13_Mapping_Earthquakes

alecngai/15_MechaCar_Statistical_Analysis

alecngai/18_Cryptocurrencies

alecngai/19_Churn_Prediction

alecngai/alecngai

Config files for my GitHub profile.

alecngai/AutoEq

Automatic headphone equalization from frequency responses

alecngai/NSUS_BI_Assessment

alecngai/Obsidian_Notebook

alecngai/Portfolio