"Boldly going where no algorithm has gone before."

A World Away: Hunting for Exoplanets with AI

S'kaiNet — Exoplanet analyzer is an AI-powered exoplanet detection system developed for the 2025 NASA Space Apps Challenge. Inspired by Star Trek’s Vulcan pursuit of knowledge — “S'kai”, meaning enlightenment — our project combines scientific curiosity with artificial intelligence to identify exoplanets from NASA’s open datasets.

Our mission: To build an intelligent platform that helps humanity discover new worlds — automatically.

NASA’s exoplanet missions — Kepler, K2, and TESS — have collected massive datasets from their space surveys. However, most exoplanet discoveries are still identified manually by scientists.

We aim to automate this process through a machine learning model trained on NASA’s open-source datasets. The model predicts whether an observation corresponds to a confirmed exoplanet, planetary candidate, or false positive — based on parameters like:

• Orbital Period
• Transit Duration
• Planetary Radius
• Stellar Brightness

1. Data Preprocessing: Clean and normalize datasets from Kepler missions.
2. Model Training: Build and tune ML models using scikit-learn.
3. Interactive UI: Develop a web interface for real-time prediction and data uploads.
4. Continuous Learning: Allow users to retrain models dynamically with new data.

Note: For images and stories created by AI, see image sources and credits in the images directory.

Dataset: Kepler Objects of Interest (KOI)

This dataset lists all confirmed exoplanets, planetary candidates, and false positives observed by the Kepler mission. We use its labeled data for supervised learning, feature optimization, and classification testing.

• ✅ Train an AI model on NASA’s exoplanet datasets
• ✅ Build a web interface for scientists and enthusiasts
• ✅ Support live visualization and accuracy tracking
• ✅ Enable users to upload and retrain data dynamically

We envision S'kaiNet — Exoplanet analyzer as a gateway for collaborative discovery — uniting data science, astronomy, and open-source innovation. Our goal is to create a platform that democratizes access to exoplanet exploration, helping both researchers and enthusiasts contribute to space science.

Our team, Outlander, reimagined our NASA Space Apps Challenge project — A World Away: Hunting for Exoplanets with AI — as a cinematic Star Trek-inspired short screenplay titled The Exoplanet Enigma.

In this fictional story, the crew of the USS Enterprise-D, led by Captain Kirk, Spock, and Number One, venture into a mysterious nebula and discover a gravitational anomaly resembling a planet. Unable to rely on visual assumptions, Spock deploys our AI-powered detection system S’kaiNet - Exoplanet analyzer, a scientific interface developed by Starfleet’s Outlander Team. Using real mission data from Kepler, the system calculates a 87 % probability that the object is a confirmed exoplanet. Through this blend of science fiction storytelling and real astrophysical methodology, The Exoplanet Enigma mirrors the same logic and data-driven exploration principles behind our actual NASA project.

🎧 The story was also adapted into an audio fiction podcast titled S’kaiNet Chronicles, combining cinematic voice acting, immersive sound design, and AI-generated narration to bring the Starfleet discovery to life.

The screenplay and podcast were built directly upon our hackathon prototype.

• The fictional S’kaiNet - Exoplanet analyzer system mirrors the architecture of our real-world AI model, which analyzes exoplanetary transit data to detect potential new worlds.
• We used OpenAI, Google Gemini, NotebookLM tools, Python, and machine-learning pipelines integrated with NASA’s exoplanet archives to train and test the model.
• The screenplay was written in Fountain format and professionally styled to standard Hollywood formatting, while the podcast was produced from the same script using AI-assisted audio synthesis and human-directed editing.

This creative adaptation demonstrates how data science, storytelling, and imagination can converge to make complex scientific concepts both accessible and inspiring.

Experience the S’kaiNet Chronicles — a fictional Starfleet log that bridges science and imagination. Follow the crew of the USS Enterprise-D as they use real NASA-inspired AI technology to uncover new worlds.

🎙️ Episode 1: The Exoplanet Enigma

• 📥 Download Audio (Right-click → Save As)
• 🎧 Listen Online (Direct stream)

🎬 Video: The Exoplanet Enigma - Visual Story

• 🎥 Download Video

📚 Comic Con Special Edition: the-exoplanet-enigma.pdf

📜 Read the full screenplay: screenplay.pdf

“Logic is the beginning of wisdom, not the end.”
— Spock