Con esta herramienta te facilitamos un acceso a todas las ofertas y demandas de tecnología europeas y a búsquedas de socios para participar en propuestas europeas de I+D publicadas en la red Enterprise Europe Network, pudiendo filtrar los resultados para facilitar las búsquedas más acordes con tus necesidades.

¿Quieres recibir estos listados de oportunidades de colaboración en tu correo de forma periódica y personalizada? Date de alta en nuestro Boletín

Para optimizar los resultados de la búsqueda, se recomienda utilizar términos en inglés.

Universidad británica ofrece un algoritmo de aprendizaje automático para identificar y agrupar características

Resumen

Tipo:
Oferta Tecnológica
Referencia:
TOUK20170109001
Publicado:
23/01/2017
Caducidad:
23/01/2018
Resumen:
Un equipo multidisciplinar de una universidad británica ha combinado técnicas de análisis de datos empleadas en astrofísica y biocomputación para desarrollar un algoritmo de aprendizaje automático capaz de identificar características en grandes conjuntos de datos y agruparlas. Este algoritmo tiene numerosas aplicaciones en el área de grandes archivos de datos. El algoritmo ha sido probado en la clasificación automática de galaxias de imágenes del telescopio espacial Hubble y actualmente la universidad está trabajando para probar el algoritmo en estudios de perfusión miocárdica y mejorar el diagnóstico de enfermedades cardíacas. La universidad busca socios con el fin de establecer acuerdos de cooperación técnica o licencia, así como inversores.

Details

Tittle:
UK university offers machine learning algorithm for feature identification and grouping seeks partners for a financial agreement, technology cooperation or licensing agreement.
Summary:
A multidisciplinary team at this UK university has united data analysis techniques used in astrophysics and bio computation to develop a machine learning algorithm that is able to identify features in large data sets and group them. The algorithm has many applications in the area of big data and the university seeks partners for technology cooperation or licensing agreements or investors for a financial agreement
Description:
The multidisciplinary team at this UK university developed the machine learning algorithm out of a need to analyse large quantities (100s Terabytes) of imaging and tabular astrophysics data. The data had to be mined and analysed to make new discoveries and the resulting data analysis tools using statistics, machine learning and modelling has a wide variety of applications in other areas that involve big data.

The main focus of the technology is in unsupervised machine learning techniques and the application of biologically-inspired machine learning algorithms for the analysis of real world data. Unsupervised learning has the advantage that it does not rely on training data but rather finds latent structure and trends in large input data sets (such as databases and images). The algorithm used has been proven in the automatic classification of galaxies in Hubble space telescope imaging and the university is currently working to prove the algorithm in myocardial perfusion scanning to improve early diagnosis of heart disease. In this regard the algorithm is very flexible and the techniques used map across a wide range of disparate fields.

The algorithm has potential in a number of areas including:

Identifying consumer patterns (e.g. supermarkets, online retail)
Predictive analysis /automatic suggestion (e.g. online retail)
Autonomous classification (e.g. medicine, defence, security)
Autonomous navigation (e.g. driverless vehicles, other automotive and aeronautical applications)

The university seeks partners in a number of areas including:
Financial agreements - the university is open to discussions with angel investors or potential spin out partners
Technology cooperation - the university would like to speak with potential technology partners to co-develop future applications
Licensing - the university is would consider licensing the knowledge and know how to suitable partners
Advantages and Innovations:
The machine learning algorithm and the computer science team that has developed this can offer:
· Bespoke solutions for complex data analysis tasks
· Access to expertise in cutting-edge data analysis and machine learning techniques
· Discovery in big data sets without the need to "teach" the system
· The ability to run the algorithm on an ordinary PC or laptop - unlike other systems that require vast computing power
Stage of Development:
Available for demonstration
IPs:
Secret Know-how,Copyright

Partner sought

Type and Role of Partner Sought:
The university seeks partners in the following areas:
Investors or companies looking for potential spin out partnerships for a financial agreement
Technology cooperation partners who would like to develop the machine learning algorithm for a particular application and would like to work with the university to do this
Licensing agreements - with companies that would like to use the technology in their products or develop solutions with it.

Client

Type and Size of Client:
University
Already Engaged in Trans-National Cooperation:
Si
Languages Spoken:
English

Keywords

Technology Keywords:
01003008 Data Processing / Data Interchange, Middleware
01003010 Databases, Database Management, Data Mining
01003012 Imaging, Image Processing, Pattern Recognition