In supervised learning tasks, a data set is available, made up of several examples for which the values of certain variables are known and the aim is to obtain a model that relates the different variables with one of them, which is the one that is to be predicted . Depending on whether the variable to be predicted is nominal (or categorical) or numerical, we speak of classification or prediction. In the Rotation Forest method, predictions are obtained by voting on several decision trees or regression. It has been used successfully in very diverse applications.
New and innovative aspects
In various fields the method has demonstrated its robustness and good results, obtaining fewer errors in the prediction than the alternatives considered.
Main advantages of its use
Prediction, whether classification or regression, is useful in countless fields. Applications where the method has already been used include diagnosis industrial, medical diagnosis, bioinformatics, performance prediction (of industrial components, students ...)
Specifications
The proposed method is based on the techniques of combining models (ensembles). The models to be combined can be classification or regression models. In particular, in Rotation Forest the combined models are decision trees or regression. It is now accepted that better results are possible combining models than using a single model, both because of the theoretical properties of these combinations and because of the wide number of applications where they have been shown to be useful.
Applications
- Industrial diagnosis. The method has been used to monitor the quality of lubricating oils, since these degrade with use. To determine the relationships between soil properties and the deterioration of mechanical pipes. To establish the state of structures.
- Prediction of returns. The method has been used to predict the power generated by wind turbines.
- Medical diagnosis. The method has been used to analyze functional magnetic resonance imaging (fMRI). These types of images are used to locate brain functions. Other applications are the diagnosis of erythematous squamous diseases or epilepsy. Also for the classification of cancer from microarrays.
- Bioinformatics. It has been used for the classification of genomic and proteomic data, for the annotation of promoter genes, for the prediction of the folding of proteins and interactions between them, prediction of molecular drug properties.
- Teaching. It has been used to identify students with learning difficulties.
- Financial. The method has been applied in predicting bankruptcies and customer risk; for municipal revenue predictio
Desired business relationship
Commercial agreement for data processing and technical studies
The aim of the The Technology Transfer Office (TTO) of the Univesidad de Burgos is to promote Innovation technology through the reseach results transfer and the conexions between the University and the new needs and requirements of the society - we are the link between the University and the Industry. Contact person: José Manuel López (jmllopez@ubu.es)
Create your free account to connect with UNIVERSIDAD DE BURGOS and thousands of other innovative organizations and professionals worldwide
Send a request for information
to UNIVERSIDAD DE BURGOS
Technology Offers on Innoget are directly posted
and managed by its members as well as evaluation of requests for information. Innoget is the trusted open innovation and science network aimed at directly connect industry needs with professionals online.
Need help requesting additional information or have questions regarding this Technology Offer?
Contact Innoget support
