Update README.md

README.md

@@ -95,24 +95,23 @@ ionicsphere_export_v7.0_*.zip/
 ### GENERATION
 The Ionic Ocean Synthetic Dataset is a specialized dataset designed to bridge the gap between complex atmospheric physics and efficient machine learning models.
 
--Dataset Overview & Purpose
-The primary goal of this dataset is to provide high-fidelity training data for neural networks to predict ionospheric conditions—specifically electron density and signal interference—without requiring the extreme computational power of traditional physics engines.
+**Goal**
+The goal of this dataset is to provide high-fidelity training data for neural networks to predict ionospheric conditions—specifically electron density and signal interference—without requiring the extreme computational power of traditional physics engines.
 
-    Target Phenomenon: It models the "Ionic Ocean," referring to the fluid-like behavior of ionized particles in the Earth's upper atmosphere (ionosphere).
-    Problem Solved: Traditional models like SAMI3 (Self-consistent Analysis of the Model of the Ionosphere) are computationally expensive and slow. This dataset allows for the training of "surrogate models" that can predict results in real-time.
-    Application: Used for improving the accuracy of GNSS/GPS positioning by predicting and correcting for atmospheric delays and signal scintillation (flickering).
+Target Phenomenon: It models the "Ionic Ocean," referring to the fluid-like behavior of ionized particles in the Earth's upper atmosphere (ionosphere).
+Problem Solved: Traditional models like SAMI3 (Self-consistent Analysis of the Model of the Ionosphere) are computationally expensive and slow. This dataset allows for the training of "surrogate models" that can predict results in real-time.
+Application: Used for improving the accuracy of GNSS/GPS positioning by predicting and correcting for atmospheric delays and signal scintillation (flickering).
 
--Key Technical Characteristics
+**Technical**
+
+-Synthetic Generation: The data is algorithmically generated, likely using a simplified physics-based simulation or a Generative Adversarial Network (GAN) to ensure it mirrors real-world radar and satellite observations.
+-Multivariate Structure: It typically contains variables representing:
+-Spatial Coordinates: Latitude, longitude, and altitude.
+-Temporal Data: Timestamps reflecting diurnal (day/night) cycles.
+-Physical Parameters: Electron density, magnetic field orientation, and solar flux indices (e.g., F10.7 index).
+-Format: Distributed as a tabular dataset (often in .csv or .parquet formats) to be compatible with common machine learning frameworks like PyTorch or TensorFlow.
 
-    Synthetic Generation: The data is algorithmically generated, likely using a simplified physics-based simulation or a Generative Adversarial Network (GAN) to ensure it mirrors real-world radar and satellite observations.
-    Multivariate Structure: It typically contains variables representing:
-        Spatial Coordinates: Latitude, longitude, and altitude.
-        Temporal Data: Timestamps reflecting diurnal (day/night) cycles.
-        Physical Parameters: Electron density, magnetic field orientation, and solar flux indices (e.g., F10.7 index).
-    Format: Distributed as a tabular dataset (often in .csv or .parquet formats) to be compatible with common machine learning frameworks like PyTorch or TensorFlow.
 
--Connection to the App
-The Ionicsphere 3D application acts as a generator for this dataset. Users can see the synthetic data points rendered as a three js 3D "ocean" surrounding the globe.
 
 ### Usage Instructions