Using auto-build toolset our data experts can quickly clean up your semi-structured data and generate models to capture complex events. Flek models are semi-supervised and when running in production can self-adjust as new data comes in.

Using our auto-discovery algorithms we can easily discover interesting strong associations, rules, bisociations, anomalies, polymalies, influencer relations as well as causal relationships between features and variables.

Thanks to the advanced prediction engine and our probabilistic methodology data science citizens can run "on-the-fly" both forward and backward predictions and make classifications without retraining or tuning. Thus, making it possible to take the best decisions given various scenarios and trends.

Flek unique probabilistic ranking algorithms can combine both profile and item based recommendation at the same time. Thus, enabling for more customization and higher conversion rate.