PROBABILISTIC GRAPHIC MODELS FOR SCALABLE DATA ANALYSIS (Q3151948): Difference between revisions
Jump to navigation
Jump to search
(Created claim: summary (P836): THE GRAPHIC MODELS PROBABILISITICOS (MGPS) HAVE UNDERGONE REMARKABLE DEVELOPMENT OVER THE PAST FEW YEARS, AND HAVE BEEN SHOWN AS VALUABLE TOOLS IN DISCIPLINES SUCH AS ARTIFICIAL INTELLIGENCE AND STATISTICS. IN THE LAST FEW YEARS, MUCH ATTENTION HAS BEEN PAID TO THE USE OF MGPS IN DATA MINING TASKS, ESPECIALLY IN SITUATIONS WITH UNCERTAINTY. In accordance with the status of the current article, the next NATURAL step is to provide them with the ca...) |
(Changed label, description and/or aliases in en: translated_label) |
||
| label / en | label / en | ||
PROBABILISTIC GRAPHIC MODELS FOR SCALABLE DATA ANALYSIS | |||
Revision as of 15:34, 12 October 2021
Project Q3151948 in Spain
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | PROBABILISTIC GRAPHIC MODELS FOR SCALABLE DATA ANALYSIS |
Project Q3151948 in Spain |
Statements
492,228.0 Euro
0 references
615,285.0 Euro
0 references
80.0 percent
0 references
1 January 2014
0 references
31 December 2017
0 references
UNIVERSIDAD DE CASTILLA-LA MANCHA
0 references
38°59'42.32"N, 1°51'21.31"W
0 references
02003
0 references
LOS MODELOS GRAFICOS PROBABILISITICOS (MGPS) HAN EXPERIMENTADO UN DESARROLLO DESTACABLE DURANTE LOS ULTIMOS AÑOS, Y SE HAN MOSTRADO COMO HERRAMIENTAS VALIOSAS EN DISCIPLINAS COMO LA INTELIGENCIA ARTIFICIAL Y LA ESTADISTICA. EN LOS ULTIMOS AÑOS, SE HA PRESTADO MUCHA ATENCION AL USO DE MGPS EN TAREAS DE MINERIA DE DATOS, ESPECIALMENTE EN SITUACIONES DOTADAS DE INCERTIDUMBRE. A PARTIR DEL ESTADO DEL ARTE ACTUAL, EL SIGUIENTE PASO NATURAL ES DOTARLOS DE LA CAPACIDAD DE OPERAR EN CONTEXTOS DE BIG DATA._x000D_ _x000D_ EL OBJETIVO PRINCIPAL DE ESTE PROYECTO ES GENERAR UN CONJUNTO DE NUEVOS DESARROLLOS METODOLOGICOS EN EL AREA DE LOS MGPS SUFICIENTEMENTE FUNDAMENTADO E INNOVADOR COMO PARA SITUARLOS DENTRO DEL AREA DEL BIG DATA COMO HERRAMIENTAS DE REFERENCIA, A TRAVES DE LA ANALITICA DE DATOS ESCALABLE. ADEMAS, EL PROYECTO PRETENDE PRODUCIR LAS HERRAMIENTAS SOFTWARE NECESARIAS PARA PERMITIR EL DESARROLLO DE APLICACIONES BASADAS EN UNA ARQUITECTURA DE SERVICIOS WEB PARA MGPS. DE ESTA FORMA, ESPERAMOS CREAR EL ENTORNO ADECUADO PARA QUE DISPOSITIVOS MOVILES PUEDAN USARSE EN CONTEXTOS DE BIG DATA, YA QUE EL NUCLEO DE LAS TAREAS DE PROCESO RECAERIAN EN UN SERVIDOR CENTRALIZADO QUE PROCESARIA LOS DATOS Y EJECUTARIA LOS ALGORITMOS, MIENTRAS QUE EL DISPOSITIVO MOVIL INTERACTUARIA A TRAVES DEL INTERFAZ DE SERVICIOS WEB. POR TANTO, EL PROPOSITO DE ESTE PROYECTO ES DOBLE, GENERANDO EN PRIMER LUGAR NUEVO CONOCIMIENTO DE LA MAS ALTA CALIDAD CIENTIFICA DENTRO DEL CAMPO DE LA ANALITICA DE DATOS ESCALABLE, PARA A CONTINUACION ABRIR EL CAMINO A UNA FRUCTIFERA TRANSFERENCIA TECNOLOGICA HACIENDO USO DE LA PLATAFORMA SOFTWARE PLANEADA._x000D_ _x000D_ LOS RESULTADOS ESPERADOS DEL PROYECTO SE PUEDEN CLASIFICAR EN CINCO CATEGORIAS:_x000D_ _x000D_ 1. MODELADO. EL PROYECTO GENERARA CONTRIBUCIONES ORIENTADAS A FORTALECER LA ESCALABILIDAD DE LOS MGPS PERMITIENDO EL ENCAPSULAMIENTO Y EL MANEJO DE DEPENDENCIAS FUNCIONALES. _x000D_ 2. INFERENCIA. SE DISEÑARAN ALGORITMOS DE INFERENCIA EFICIENTES Y ESCALABLES, TOMANDO COMO BASE LOS ARBOLES DE PROBABILIDAD RECURSIVOS._x000D_ 3. APRENDIZAJE. SE DISEÑARAN ALGORITMOS ESCALABLES DE APRENDIZAJE TENIENDO EN CUENTA LAS RESTRICCIONES DE OPERAR EN ENTORNOS DE BIG DATA. ESTO INCLUIRA MODELOS CANONICOS Y APRENDIZAJE A PARTIR DE STREAMS DE DATOS. TAMBIEN SE DESARROLLARAN ALGORITMOS NO-ESTANDARES DE CLASIFICACION CAPACES DE APRENDER, POR EJEMPLO, A PARTIR DE CONJUNTOS DE DATOS CON MULTIPLES ETIQUETAS._x000D_ 4. SOFTWARE. SE IMPLEMENTARA UNA PLATAFORMA SOFTWARE Y SE PONDRA A DISPOSICION DE LA COMUNIDAD, DONDE SE INCLUIRAN LOS ALGORITMOS DESARROLLADOS EN LAS TAREAS METODOLOGICAS Y SE POSIBILITARA EL DESARROLLO DE APLICACIONES A TRAVES DE UN INTERFAZ DE SERVICIOS WEB._x000D_ 5. APLICACIONES. SE ABORDARA UN AMPLIO ABANICO DE APLICACIONES, CON OBJETO DE APLICAR LAS PROPUESTAS METODOLOGICAS DEL PROYECTO. SE CUBRIRAN LAS AREAS DE AVIACION, SALUD, MULTIMEDIA Y FINANZAS._x000D_ _x000D_ EN ESTE SUBPROYECTO NOS CENTRAREMOS EN LOS ASPECTOS DE APRENDIZAJE SUPERVISADO Y NO SUPERVISADO, Y APLICACIONES A MULTIMEDIA Y BIO-MEDICINA. (Spanish)
0 references
THE GRAPHIC MODELS PROBABILISITICOS (MGPS) HAVE UNDERGONE REMARKABLE DEVELOPMENT OVER THE PAST FEW YEARS, AND HAVE BEEN SHOWN AS VALUABLE TOOLS IN DISCIPLINES SUCH AS ARTIFICIAL INTELLIGENCE AND STATISTICS. IN THE LAST FEW YEARS, MUCH ATTENTION HAS BEEN PAID TO THE USE OF MGPS IN DATA MINING TASKS, ESPECIALLY IN SITUATIONS WITH UNCERTAINTY. In accordance with the status of the current article, the next NATURAL step is to provide them with the capacity to operate in connection with BIG DATA._x000D_ _x000D_ The main objective of this project is to achieve a new set of methodological developments in the AREA of the MGPS. funded and INNOVATING as to place them within the AREA of the BIG DATA as REFERENCE TOOLS, through the analysis of ESCALABLE DATA. IN ADDITION, THE PROJECT AIMS TO PRODUCE THE SOFTWARE TOOLS NEEDED TO ENABLE THE DEVELOPMENT OF APPLICATIONS BASED ON A WEB SERVICE ARCHITECTURE FOR MGPS. IN THIS WAY, WE HOPE TO CREATE THE RIGHT ENVIRONMENT FOR MOBILE DEVICES TO BE USED IN BIG DATA CONTEXTS, AS THE NUCLEUS OF THE PROCESS TASKS RECAERIAN ON A CENTRALISED SERVER THAT WOULD PROCESS THE DATA AND EXECUTE THE ALGORITHMS, WHILE THE MOBILE DEVICE WOULD INTERACT THROUGH THE WEB SERVICE INTERFACE. Therefore, the PROPOSITE OF THIS PROJECT is DOUBLE, GENERAN IN THE FIRST PLACE NEW KNOWING OF THE HIGH HIGH QUALITY SCIENTIFIC QUALITY DATES ESCALABLE ANALYTICAL, FOR CONTINUATION TO open the door to a fruitful TRANSFERENCE TECHNOLOGICAL FROM THE SOFTWARE PLATAFORM PLANEADA._x000D_ _x000D_ SPERED RESULTS OF THE PROJECT may be classified in FIVE CATEGORIES:_x000D_ _x000D_ 1. MODELING. THE PROJECT WILL GENERATE CONTRIBUTIONS AIMED AT STRENGTHENING THE SCALABILITY OF THE MGPS BY ALLOWING THE ENCAPSULATION AND MANAGEMENT OF FUNCTIONAL UNITS. _x000D_ 2. INFERENCE. Efficient and scalable INFERENCE ALGORITMS will be designed, taking as a basis the RECURSIVE PROBABILITY ARBOLES._x000D_ 3. LEARNING. SCALABLE LEARNING ALGORITHMS WILL BE DESIGNED TAKING INTO ACCOUNT THE CONSTRAINTS OF OPERATING IN BIG DATA ENVIRONMENTS. THIS WILL INCLUDE CANONICAL MODELS AND LEARNING FROM DATA STREAMS. Non-Standard CLASSIFICATION ALGORITS will also be developed._x000D_ 4. SOFTWARE. A SOFTWARE PLATAFORM will be implemented and made available to the Community, WHERE THE ALGORITMES developed in the METODOLOGICAL TARES will be included and the development of applications made possible through an interfacing of web services._x000D_ 5. APPLICATIONS. A WIDE RANGE OF APPLICATIONS WILL BE ADDRESSED IN ORDER TO IMPLEMENT THE PROJECT’S METHODOLOGICAL PROPOSALS. Aviation, HEALTH, MULTIMEDIA AND FINANZAS AREAS._x000D_ _x000D_ on this sub-project we will focus on SUPERVISED AND NON-SUPERVISED LEARNING ASPECTS, AND MULTIMEDIA and BIO-MEDICINE APPLICATIONS. (English)
12 October 2021
0 references
Albacete
0 references
Identifiers
TIN2013-46638-C3-3-P
0 references