Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/LauraSilRu/exoplanet-profiler/llms.txt

Use this file to discover all available pages before exploring further.

ExoProfiler’s clustering pipeline applied to 731 TESS exoplanets identifies two stable, interpretable families. The Standard Population accounts for 91.24% of the sample and captures the mainstream of rocky and gaseous mid-size worlds. The Exotic group (8.76%) isolates massive gas giants and Hot Jupiters with extreme orbital properties.

Family Overview

Familia 0 — Standard Population

667 exoplanets · 91.24% of sampleRocky, Super-Earths, Sub-Neptunes — the dominant regime of TESS discoveries.

Familia 1 — Exotic Objects

64 exoplanets · 8.76% of sampleMassive Gas Giants, Hot Jupiters — extreme orbital and physical parameters.

Cluster Structure Diagram

The K-Means algorithm partitions all 731 TESS exoplanets into two groups at a clean decision boundary in 4-dimensional PCA space.
              ┌───────────────────────────┐
              │      731 EXOPLANETS       │
              └─────────────┬─────────────┘

            ┌───────────────┴─────────────────┐
            ▼                                 ▼
      ┌─────────────────┐             ┌─────────────────┐
      │    CLUSTER 0    │             │    CLUSTER 1    │
      │  667 exoplanets │             │   64 exoplanets │
      │    (91.24%)     │             │     (8.76%)     │
      └────────┬────────┘             └────────┬────────┘
               │                               │
               ▼                               ▼
      "Standard / Mainstream"         "Exotic / Extreme"

Family 0 — Standard Population

  • Size: 667 exoplanets (91.24%)
  • Character: Covers the continuum from terrestrial rocky planets through Super-Earths to Sub-Neptunes.
  • Orbital Properties: Moderate orbital periods, compact semi-major axes — these are typically the “rule” for TESS detections.
  • Astrophysical Analogs: Corresponds to NASA categories: Terrestrials (< 2 R⊕), Super-Earths (1.5–2 R⊕), Sub-Neptunes (2–4 R⊕).
The transit photometry method used by TESS has a detection bias toward larger planets in close orbits, which may influence which planets populate this cluster.

Family 1 — Exotic Objects

  • Size: 64 exoplanets (8.76%)
  • Character: Gas Giants and Hot Jupiters with extreme orbital or physical parameters.
  • Orbital Properties: Some have wide orbits with high eccentricity; others are scorching Hot Jupiters on ultra-close orbits.
  • Astrophysical Analogs: Corresponds to Gas Giants / Hot Jupiters (> 4 R⊕, high mass).
DBSCAN identifies this population as statistical outliers relative to the main density core — interpret individual planets in this group as potentially anomalous.

Astrophysical Interpretation

The two families are data-driven mathematical structures, not official astronomical classifications. They broadly align with well-known NASA planet type groupings, but the boundary between families is defined by the K-Means decision surface in 4-dimensional PCA space, not by fixed physical thresholds.

Column Definitions

The output file clustered_exoplanets.csv contains the following columns:
ColumnDescription
pl_namePlanet identifier
PC1First principal component score
PC2Second principal component score
PC3Third principal component score
PC4Fourth principal component score
Cluster_K2Integer label: 0 = Standard Population, 1 = Exotic Objects
Familia_PlanetaDescriptive string label for the assigned family

Loading the Results

import pandas as pd

df = pd.read_csv('data/processed/clustered_exoplanets.csv')

# View family distribution
print(df['Familia_Planeta'].value_counts())

# Access a specific family
exotics = df[df['Cluster_K2'] == 1]
standard = df[df['Cluster_K2'] == 0]

Build docs developers (and LLMs) love