We build predictive models, computer vision, natural language processing, intelligent agents that uplift your business potential.
Molecular properties prediction
Molecular machine learning has been maturing rapidly over the last few years. Improved methods and the presence of larger datasets have enabled machine learning algorithms to make increasingly accurate predictions about molecular properties. Those may vary from physical-chemical properties, ADME, toxicity, up to binding affinity to protein targets. We specialized such systems to predict the targets desired by our customers.
Trade flows analysis
Trade flows don’t tell the whole story when trying to determine the ultimate source of products, because of re-imports(exports) and product transformations. Through input-output analysis, it is possible to disentangle the loops and knots of the trade flows and retrace the original source of products.
Tumor risk assessment
We developed a predictive model that, based on the result of multiple diagnostic tools, is able to evaluate prostate cancer risk. This model proved to outperform currently used diagnostic approaches.
Marketing for retailers
We developed a customer profiling system for retailers. This profiling system is coupled with a recommender and forecasting model that allows the retailer to optimally sell their products.
We used deep survival analysis techniques to better determine the risk of developing a number of diseases on the basis of clinical data and patient's reported lifestyle.