Causation-based machine learning for early and accurate assessment of a situation.
Utilizing artificial intelligence for early detection in geological, chemical, and biological systems remains a struggle for multiple reasons.
- There are limited examples of events. This results in too little data to train a model.
- There is no clear indication before an event or state change. Interpretation isn't achievable with statistical models.
- Significant delay between a cause and the resulting effect prevents models from seeing correlation. This is critical due to the catalytic properties present in nature.
- Probabilistic answers aren’t sufficient. Definitive answers are needed to forewarn of events where lives or significant dollars are at risk.
Intuition AI was developed with a data-light approach, making definitive predictions with a handful of data sets.
Unlike traditional machine learning, the Senslytics model combines domain expert knowledge with dynamically selected data allowing for understanding of all the situational factors.
Senslytics software can identify the cause of an event using situation signatures and validates data through multiple views. It then generates conclusions that are more definitive any than probabilistic approach.
Read our full Technical Overview.
Unlike most machine learning applications, Senslytics’ Intuition AI platform is:
- Definitive: the model knows when it has a reliable interpretation and when it does not
- Explainable: reasoning is provided for all decisions
- All-Encompassing: understands "outlier" behavior
Use Case
Mudgas Reservoir Estimation
Move mudgas logging estimations closer to ground truth for GOR, viscosity, API gravity, and density.
Use Case
Corrosion Monitoring
Identify areas of accelerated corrosion growth early to increase asset run time and longevity.
Use Case
Contamination Monitoring
Avoid fluid sample contamination and excessively long jobs during wireline formation tests.
