Repository containing my portfolio of data science and cybersecurity projects completed in iPython Notebooks, R Markdown files, Tableau, Power BI, SQL files and Azure ML for learning and hobby purposes.
Author: My Pham
Feel free to copy and refer to my repos. If you have any questions, please contact me on my LinkedIn or at [email protected].
Please also visit my other portfolios for visualization and model automation projects:
Python
- Supervised Learning - Stock Price Prediction: a combination of regression models (time series, regression machine learning, ensemble learning and deep learning) to provide financial investors insights on a diversified portfolio of 6 companies representing 6 different industries and the S&P500 index and forecast stock prices of each company to investors make financial decisions based on the price movements
- Supervised Learning - Heart Disease Prediction: a mix of classification models (ensemble learning, classification machine learning, principal component analysis and deep learning) to provide hospitals a filtering system that helps patients identify and prevent potential predictors of having a heart disease
- Supervised Learning - Bike Rental Prediction: a combination of regression models (ensemble learning, classification machine learning, principal component analysis and deep learning) to predict the demand for bike rental based on a number of attributes, projected to help the company effectively manage budgets and provide enough supply to meet the predicted demand
Tools: pandas, numpy, matplotlib, seaborn, tensorflow, keras, beautifulsoup, requests, scikit-learn, statsmodels.tsa, mysql.connector
Azure ML
- Stock Price Prediction Using ARIMA Time Series: part of Stock Price Prediction, see full project above
- Income Classification Using Logistic Regression: a model using Adult Census Income Data to predict whether a person earns more than $50,000/year or not based on 15 attributes with logistic regression method.
Python
- Unsupervised Learning - Yelp's Reviews Classification Using NLP Text Analytics: an analysis of Yelp's reviews using NLP Text Analytics to classify whether a review is positive or negative, helping businesses determine areas of improvement and create strategic plans accordingly
- Fake News Detection: a filtering program using NLP Text Analytics to detect fake news tat negatively impacts businesses and stock market
*Tools: pandas, numpy, matplotlib, nltk, matplotlib, seaborn, string, itertools, requests, Counter, Text, word_tokenize, sent_tokenize, regexp_tokenize, stopword, Dictionary, TfidfModel, scikit-learn, wordcloud
MySQL
- Stock Market Analysis Using MySQL: part of Stock Price Prediction, see full project above
- Google Analytics Customer Revenue Prediction Using MySQL: an analysis of Google's visitor data to help companies using this data develope plans to save marketing budgets and increase revenue by identifying trends of visitors' usage and behavior
Tableau
- Stock Market Analysis Using Tableau: part of Stock Price Prediction, see full project above
Power BI
- Payment Cards Anomaly Detection Using Power BI: an analysis of payment cards of a fictitious company using Power BI visualization to identify unusual transactions or suspicious activities, helping the accounts payable, accounts receivable and audit departments have appropriate internal controls in place to prevent potential fraud
Python
- 100 Days Of Code: Python projects from beginner level to advanced level in 100 days
- Fraud Detection Using Audit Technologies (ACL, MS Excel, IDEA): audit programs to detect fraud using audit technologies (ACL, MS Excel, IDEA), projected to reduce audit process duration and costs and manage transactions more effectively
- Information Technology Cyber Security Policy Proposal For Venmo: Information technology security policy proposal For Venmo including vulnerability assessment, risk assessment, control analysis and recommendations to prevent cyber threats
- Information Technology Audit Reports: a IT policy report containing risk assessment, risk matrix and control recommendations to mitigate risk of fraud
- Fake News Detection: see full project above
- Payment Cards Anomaly Detection Using Power BI: see full project above