Artificial Organic Networks (AONs) is a new artificial intelligence technique bio-inspired in organic chemistry firstly proposed by Hiram Ponce and Pedro Ponce, in 2010. This research presents an alternative technique for improving two key features in computational algorithms for modeling problems: stability in algorithms and partial knowledge of model behavior.

Computational algorithms for modeling problems are widely used in real applications, such as: predicting and describing systems, finding patterns on unknown or uncertain data. Interesting applications have been developed for engineering, biomedics, chemistry, economics, physics, statistics, and so on.

However, the related computational algorithms have some drawbacks. For instance, several methods assume partial knowledge of systems, others suppose specific dependeces on variable instances, some others act as black boxes without concerning on how models work, like other methods are not stable due to heuristics or arbitrary parameters difficult for setting.