Adding earthquake prediction model

Pull Request Description: Earthquake Prediction Overview: This pull request adds a comprehensive Earthquake Prediction project utilizing the dataset /kaggle/input/the-ultimate-earthquake-dataset-from-1990-2023/Eartquakes-1990-2023.csv from Kaggle. The project employs machine learning models to predict seismicity levels (low, medium, high) based on earthquake parameters spanning from 1990 to 2023. Changes Made:

Notebook: • Implemented Earthquake_Prediction.ipynb: This notebook integrates the following key components: o Exploratory Data Analysis (EDA): Investigates the dataset's characteristics using histograms, scatter plots, count plots, and correlation heatmaps. o Data Preprocessing: Creates a categorical target variable ('seismicity') based on earthquake magnitudes, bins the magnitudes into predefined categories, and handles missing data. o Modeling and Evaluation: Implements three classifiers:  Naive Bayes Classifier (GaussianNB): Performs classification based on Gaussian distribution assumptions.  Support Vector Machine (SVM): Applies a linear kernel for classification.  Random Forest Classifier: Utilizes an ensemble of decision trees for classification. o Model Evaluation: Assesses model performance using accuracy scores, classification reports, and confusion matrices. o Visualization: Includes visual representations such as scatter plots, histograms, and heatmaps to interpret data and model outcomes.

Dataset: • Included /kaggle/input/the-ultimate-earthquake-dataset-from-1990-2023/Eartquakes-1990-2023.csv: This dataset contains comprehensive earthquake data attributes necessary for the project, including magnitude, location coordinates, depth, tsunami occurrence, and significance. Summary of Analysis: • Conducted thorough exploratory analysis to understand earthquake data patterns and correlations. • Trained multiple machine learning models to predict seismicity levels based on significant earthquake parameters. • Evaluated model performance using standard metrics and visualized results to enhance interpretability and insights. Additional Notes: • The notebook is structured for clarity and includes detailed explanations of each step and visualization. • Dataset integrity and completeness have been ensured for reliable analysis and model training.

Screenshots: •

Checklist:

[x] Added Jupyter notebook (Earthquake_Prediction.ipynb) for earthquake prediction project.
[x] Included dataset (Eartquakes-1990-2023.csv) from Kaggle for analysis.
[x] Conducted comprehensive EDA and visualization of earthquake data.
[x] Implemented and evaluated machine learning models for seismicity prediction.
[x] Ensured clear documentation and explanations within the notebook for ease of understanding.

Recode-Hive / machine-learning-repos

Adding earthquake prediction model #469