Using our tools for building machine learning models, we deploy solutions across multiple industries. Our algorithms are data agnostic, which provides a scalable approach to solving many of the world’s toughest problems.
Fault detection on an intensified heat-exchanger reactor
Process monitoring and fault diagnosis are of great importance for operation safety and efficiency of complex industrial plants. MinerWorx was used in this case to address the sensor location problem for fault detection. All the process situations were identified using the whole available set of sensors and then an optimal set of sensors that characterize accurately different situations (abnormal and normal) were selected.
Breast Cancer Prognosis
Breast cancer prognosis using original set of biomarkers and data analysis software based on fuzzy logic method. In oncology one of greatest subjects of study is how to better identify low risk recurrence population avoiding therefore chemotherapy and its secondary effects. From beginning of 2000’s new biomarkers have been selected by molecular methods in order to improve the tumor prognosis.Worldwide investment resources are dedicated to seek new efficient biomarkers able to better appreciate the risk of recurrence prognosis.
Optimize and predict the success of future promotional operations
Manufacturers of consumer goods are subject to strong pressure from distributors and consumers to multiply promotional offers. Effective management of their promotional operations has become vital. MinerWorx has been used in this case to build a predictive model that can predict future situations, such as the success of a promotional operation.