AI workflow classifies subsurface geology using wireline logs in Ghana

Researchers have developed an unsupervised machine learning approach to classify rock formations and estimate porosity in the Keta Basin, Ghana, using only wireline log data. The method applied K-means clustering to analyze six different wireline logs from a specific well, identifying four distinct electrofacies. These identified clusters correlate with variations in clay content and rock properties, offering a practical tool for subsurface characterization in areas lacking core data. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Provides a log-only, unsupervised clustering framework for subsurface characterization in frontier basins.

RANK_REASON This is a research paper published on arXiv detailing a new methodology for geological analysis.

Read on arXiv cs.AI →

• Keta Basin
• Ghana
• K-means clustering
• arXiv

• paper
• other