Tag: data science

  • Prediction of NASA Turbofan Jet Engine RUL: OLS, SciKit-Learn & LSTM

    Prediction of NASA Turbofan Jet Engine RUL: OLS, SciKit-Learn & LSTM

    We predict the Remaining Useful Life (RUL) of NASA turbofan jet engines by comparing the statsmodels OLS, ML SciKit-Learn regression vs LSTM Keras in Python. The input dataset is the Kaggle version of the public dataset for asset degradation modeling from NASA. It includes Run-to-Failure simulated data from turbo fan jet engines.

  • The 5-Step GCP IoT Device-to-Report via AI Roadmap

    The 5-Step GCP IoT Device-to-Report via AI Roadmap

    The Internet of Things (IoT) aids in the improvement of processes and enables new scenarios through network-connected devices. Recognized as a driver of the Fourth Industrial Revolution, IoT applications include predictive maintenance, industry safety, automation, remote monitoring, asset tracking, and fraud detection. Advancements in cloud IoT architectures over recent years have enabled efficient data ingestion,…

  • Health Insurance Cross Sell Prediction with ML Model Tuning & Validation

    Health Insurance Cross Sell Prediction with ML Model Tuning & Validation

    The content discusses the use of AI and Machine Learning (ML) for insurance cross-selling. It covers topics such as data preparation, model training with different algorithms, parameter optimization, and model evaluation. The study showcases the ability of ML models (HGBM, XGBoost, Random Forest) to predict cross-sell customers in the insurance sector, providing potential for improved…

  • Plotly Dash TA Stock Market App

    Plotly Dash TA Stock Market App

    The post explains how to deploy a Plotly Dash stock market app in Python with the dashboard of user-defined stock prices. This includes technical indicators like volume, MACD, and stochastic. The steps include selecting a stock ticker symbol (NVDA), retrieving stock data from yfinance API, adding Moving Averages, saving the stock chart in HTML form,…

  • Low-Code AutoEDA of Dutch eHealth Data in Python

    Low-Code AutoEDA of Dutch eHealth Data in Python

    The article details the usage of Python’s Low-Code AutoEDA for examining Dutch Healthcare Authority’s eHealth data. Utilizing various Python libraries like D-Tale, SweetViz, etc., the study aims to understand the healthcare data’s key features to ready it for AI techniques. The motivations include the Dutch government’s support for digital healthcare applications, especially amidst the recent…

  • NVIDIA Rolling Volatility: GARCH & XGBoost

    NVIDIA Rolling Volatility: GARCH & XGBoost

    This post examines the prediction of NVIDIA stock volatility using two models: the Generalized Autoregressive Conditional Heteroscedasticity (GARCH) and the Extreme Gradient Boosting (XGBoost). Both models are compared in terms of MSE and MAPE. The post discovers that the machine learning-based XGBoost model outperforms the GARCH model in NVDA volatility forecasting, showing the effectiveness of…

  • Time Series Forecasting of Hourly U.S.A. Energy Consumption – PJM East Electricity Grid

    Time Series Forecasting of Hourly U.S.A. Energy Consumption – PJM East Electricity Grid

    Table of Contents PJME Data Let’s set the working directory YOURPATH and import the following key libraries Let’s read the input csv file in our working directory Let’s plot the time series Data Preparation Output: (113926, 1, 9) (113926,) (31439, 1, 9) (31439,) LSTM TSF Let’s plot the LSTM train/test val_loss history Output: MSE: 1811223.125…

  • Wind Energy ML Prediction & Turbine Power Control

    Wind Energy ML Prediction & Turbine Power Control

    This text presents a detailed project on modeling the power curve of a wind turbine, which is crucial in wind energy management and forecasting. By using machine learning techniques such as Random Forest and Gradient Boosting Regressors, and validating with real-world Scada data from a Turkish wind farm, the project shows it’s possible to create…

  • Image Based Fast Forest Fire Detection with TensorFlow

    Image Based Fast Forest Fire Detection with TensorFlow

    A recent study showcases the use of artificial intelligence (AI) and deep learning (DL) for efficient wildfire prediction and management. Utilizing a fast DL approach based on the TensorFlow Convolution Neural Network (CNN) algorithm, researchers trained models to distinguish between fire and non-fire images using a public-domain dataset. The implemented system predicted fires accurately and…

  • Morocco Earthquake EDA

    Morocco Earthquake EDA

    Featured design via Canva. Clickable Table of Contents Basic Installations and Imports Let’s set the working directory YOURPATH Let’s install and import the following libraries Download Earthquake Input Data For this project, we’ll use a dataset that contains all seismic events over the last seven days, which have a magnitude of 1.0 or greater: Output:…

  • EUR/USD Forecast: Prophet vs JPM

    EUR/USD Forecast: Prophet vs JPM

    JP Morgan (JPM) analysis predicts the EUR/USD exchange rate to hold at 1.08 in December 2023, while ING forecasts suggest rates of $1.00 throughout 2023 and $1.02 in Q1 2024, rising to $1.10 by Q4 2024. Using the FB Prophet model, predictions show a hold at 1.08 +/- 0.07 in December 2023, aligning with JPM’s…

  • WA House Price Prediction: EDA-ML-HPO

    WA House Price Prediction: EDA-ML-HPO

    A predictive model of house sale prices in King County, Washington, was developed using multiple supervised machine learning (ML) regression models, including LinearRegression, SGDRegressor, RandomForestRegressor, XGBRegressor, and AdaBoostRegressor. The best-performing model, XGBRegressor, explained 90.6% of the price variance, with a RMSE of $18472.7. These results, valuable to local realtors, indicate houses with a waterfront are…

  • ML Prediction of High/Low Video Game Hits with Data Resampling and Model Tuning

    ML Prediction of High/Low Video Game Hits with Data Resampling and Model Tuning

    The post outlines a ML-based approach to forecast video game sales, using several techniques to enhance training, accuracy, and prediction. The Kaggle’s VGChartz dataset, containing sales data and other game-specific information, was used to build and refine the model. Several ML techniques including RandomForestClassifier and Logistic Regression yielded top predictors, with the critic’s score deemed…

  • Improved Multiple-Model ML/DL Credit Card Fraud Detection: F1=88% & ROC=91%

    Improved Multiple-Model ML/DL Credit Card Fraud Detection: F1=88% & ROC=91%

    In 2023, the global card industry is projected to suffer $36.13 billion in fraud losses. This has necessitated a priority focus on enhancing credit card fraud detection by banks and financial organizations. AI-based techniques are making fraud detection easier and more accurate, with models able to recognize unusual transactions and fraud. The post discusses a…

  • Datapane Stock Screener App from Scratch

    Datapane Stock Screener App from Scratch

    This content provides a quick guide for value investors to use the Datapane stock screener API in Python. It includes instructions for installation, importing standard libraries, setting the stock ticker, downloading stock Adj Close price, and creating visualizations. The post also describes how to build a powerful report using Datapane’s layout components.

  • Unsupervised ML, K-Means Clustering & Customer Segmentation

    Unsupervised ML, K-Means Clustering & Customer Segmentation

    Table of Clickable Contents Motivation Methods Open-Source Datasets This file contains the basic information (ID, age, gender, income, and spending score) about the customers. Online retail is a transnational data set which contains all the transactions occurring between 01/12/2010 and 09/12/2011 for a UK-based and registered non-store online retail. The company mainly sells unique all-occasion…

  • Working with FRED API in Python: U.S. Recession Forecast & Beyond

    Working with FRED API in Python: U.S. Recession Forecast & Beyond

    The FRED API, or Federal Reserve Economic Data, provides over 267,000 economic time series from 80 sources, offering a wealth of data to promote economic education and research. It encompasses U.S. economic and financial data, including interest rates, monetary indicators, exchange rates, and regional economic data. Additionally, we analyzed correlations, trained currency exchange prediction models,…

  • Dabl Auto EDA-ML

    Dabl Auto EDA-ML

    Dabl, short for Data Analysis Baseline Library, is a high-level data exploration library in Python that automates repetitive data wrangling tasks in the early stages of supervised machine learning model development. Developed by Andreas Mueller and the scikit-learn community, it facilitates data preprocessing, advanced integrated visualization, exploratory data analysis (EDA), and ML model development, demonstrated…

  • Using AI/ANN AUC>90% for Early Diagnosis of Cardiovascular Disease (CVD)

    Using AI/ANN AUC>90% for Early Diagnosis of Cardiovascular Disease (CVD)

    The project utilizes AI-driven cardiovascular medicine with a focus on early diagnosis of heart disease using Artificial Neural Networks (ANN). Aiming to improve early detection of heart issues, the project processed a dataset of 303 patients using Python libraries and conducted extensive exploratory data analysis. A Sequential ANN model was subsequently built, revealing excellent performance…

  • JPM Breakouts: Auto ARIMA, FFT, LSTM & Stock Indicators

    JPM Breakouts: Auto ARIMA, FFT, LSTM & Stock Indicators

    The post discusses predicting JPM stock prices for 2022-2023 using several predictive models like ARIMA, FFT, LSTM, and Technical Trading Indicators (TTIs) such as EMA, RSI, OBV, and MCAD. The ARIMA model used historical data, while the partial spectral decompositions of stock prices served as features for the FFT model. TTIs were calculated to validate…