There is a total for 5 projects done in this course.
The following are the brief summary of the projects.
01 : Can Standardized Test Be Used As The Only Way To Reduce Drop Out?
• Brief Summary :This is the first project done find out if standardized test can be used as the only way to reduce drop out.
• Libraries used : Python, Jupyter Notebook, Pandas, Matplotlib, Numpy
02 : Required core features for the next development project
• Brief Summary : Analyse the factors that affect the house price and come out with core factors to focus on to improve sales price with 84% accuracy score.
• Libraries used : Python, Jupyter Notebook, Pandas, Matplotlib, Seaborn, Shap, Sklearn, Ipython.display, Numpy
• Machine Learning models used : Linear regression, Lasso Regression, Ridge Regression
03 : Reddit Classification : Depression vs Anxiety
• Brief Summary : Predict which words belongs to which subreddit with up to 96% accuracy score, 96% precise score and 96% recall score.
• API used : Pushshift Reddit (REST API)
• Libraries used : Python, Jupyter Notebook, BeautifulSoup, Pandas, Matplotlib, Request, Datetime, Time, Random, String, Re, Seaborn, Sklearn
• Natural Language Processing (NLP) used : Count Vectorizer, TF-IDF, Spacy, WordCloud, NLTK
• Machine Learning models used :Logistic Regression, Naïve Bayes Bernoulli, Random Forest, Adaboost, Gradient Boosting
04 : West Nile Prediction
• Brief Summary : Predict the virus location with up to 93% recall score with demo to show the locations of the virus location.
• Libraries used : Python, Jupyter Notebook, Pandas, Matplotlib, Sklearn, ploty, datetime, OneHotEncoder, RobustScaler, Smote, StratifiedKFold
• Machine Learning models used : Logistic Regression, Random Forest, Adaboost
• Demo used : Streamlit
05 : Crypto (Eth) Fraud Detection
• Brief Summary : Developed a computational efficient (62% reduction in features) model for fraud detection in Ethereum with 97% recall score, 97% precise score and 95% accuracy score.
• Libraries used : Python, Jupyter Notebook, Pandas, Matplotlib, Sklearn, GridsearchCV, Smote, Seaborn
• Machine Learning models used : Logistic Regression, Random Forest, Adaboost, Xgboost