Search results for "LEARNING"

showing 10 items of 6669 documents

A recommendation system for the prediction of drug-target associations

2022

In this chapter a recommendation system is presented, based on the integration of a Protein-Protein Interaction (PPI) network taken from the Intact database, and a set of associations between drugs and targets taken from the DrugBank database. Depending on how proteins are connected on the PPI network, given an input drug the system suggests new targets. The framework adopted for the implementation is Apache Spark, useful for loading, managing and manipulating data by means of appropriate Resilient Distributed Datasets (RDD), and for the use of the Alternating Least Square (ALS) machine learning algorithm, a Matrix Factorization algorithm for distributed and parallel computing.Finally, an a…

machine learningSettore INF/01 - Informaticacollaborative filteringdrugs
researchProduct

Impact of Renewables in Spanish Electricity Markets

2021

El objetivo de la presente tesis es analizar el impacto de la introducción de las energías renovables en el sistema eléctrico español. El sistema eléctrico español es elegido como ejemplo paradigmático debido al intenso crecimiento observado en las renovables en los últimos años, en especial de energía eólica. Su contenido se estructura en 3 capítulos: En el primer Capitulo, ”Effects of renewable on the stylized facts of electricity prices”, se analiza el impacto de las renovables en el precio resultante de la subasta del mercado diario (también llamado precio spot) siendo objeto de estudio no solo el comportamiento del precio en niveles, sino también sus características principales, como …

machine learningbalancing marketsintermittencyelectricity marketstrategic biddingUNESCO::CIENCIAS ECONÓMICASrenewable:CIENCIAS ECONÓMICAS [UNESCO]price volatility
researchProduct

Exploring relationships between audio features and emotion in music

2009

In this paper, we present an analysis of the associations between emotion categories and audio features automatically extracted from raw audio data. This work is based on 110 excerpts from film soundtracks evaluated by 116 listeners. This data is annotated with 5 basic emotions (fear, anger, happiness, sadness, tenderness) on a 7 points scale. Exploiting state-of-the-art Music Information Retrieval (MIR) techniques, we extract audio features of different kind: timbral, rhythmic and tonal. Among others we also compute estimations of dissonance, mode, onset rate and loudness. We study statistical relations between audio descriptors and emotion categories confirming results from psychological …

machine learningclassificationaudioemotionmirmusic
researchProduct

Impacto de una nueva metodología de análisis de biomarcadores de imagen ecográfica en ciclos sustituidos sobre la tasa de éxito en donación de óvulos

2019

En las últimas décadas la imagen médica se ha convertido en una herramienta indispensable para conocer y analizar el interior del cuerpo humano de forma no invasiva. Además, la digitalización de la imagen y el desarrollo de técnicas de procesado computacional permiten en la actualidad extraer información contenida en la imagen imperceptible para el ojo humano. Esta información extraída de la imagen se conoce como Biomarcador de Imagen (BI). Un Biomarcador de Imagen (BI) se define como un parámetro que representa y cuantifica de forma objetiva una propiedad (estructural, funcional o biológica) del tejido objeto de estudio, y que se comporta como indicador de un proceso biológico normal, una …

machine learningembarazointeligencia artificialUNESCO::CIENCIAS MÉDICASimplantación embrionariadonación de óvulosanálisis de textura:CIENCIAS MÉDICAS [UNESCO]ecografía
researchProduct

Transfer Learning of Deep Learning Models for Cloud Masking in Optical Satellite Images

2023

Los satélites de observación de la Tierra proporcionan una oportunidad sin precedentes para monitorizar nuestro planeta a alta resolución tanto espacial como temporal. Sin embargo, para procesar toda esta cantidad creciente de datos, necesitamos desarrollar modelos rápidos y precisos adaptados a las características específicas de los datos de cada sensor. Para los sensores ópticos, detectar las nubes en la imagen es un primer paso inevitable en la mayoría de aplicaciones tanto terrestres como oceánicas. Aunque detectar nubes brillantes y opacas es relativamente fácil, identificar automáticamente nubes delgadas semitransparentes o diferenciar nubes de nieve o superficies brillantes es mucho …

machine learningflood detectioncloud maskingtransfer learningUNESCO::CIENCIAS TECNOLÓGICAS
researchProduct

Towards a theory of inductive inference

1973

machine learninginductive inference
researchProduct

Multilayer perceptron training with multiobjective memetic optimization

2016

Machine learning tasks usually come with several mutually conflicting objectives. One example is the simplicity of the learning device contrasted with the accuracy of its performance after learning. Another common example is the trade-off that must often be made between the rate of false positive and false negative predictions in diagnostic applications. For computer programs that learn from data, these objectives are formulated as mathematical functions, each of which describes one facet of the desired learning outcome. Even functions that intend to optimize the same facet may behave in a subtly different and mutually conflicting way, depending on the task and the dataset being examined. Mul…

machine learningkoneoppiminenclassification algorithmsmemeettiset algoritmitalgoritmitmultiobjective optimizationmultilayer perceptronmemetic algorithmsneuroverkotmatemaattinen optimointineural networksluokitus
researchProduct

Node co-activations as a means of error detection : Towards fault-tolerant neural networks

2022

Context: Machine learning has proved an efficient tool, but the systems need tools to mitigate risks during runtime. One approach is fault tolerance: detecting and handling errors before they cause harm. Objective: This paper investigates whether rare co-activations – pairs of usually segregated nodes activating together – are indicative of problems in neural networks (NN). These could be used to detect concept drift and flagging untrustworthy predictions. Method: We trained four NNs. For each, we studied how often each pair of nodes activates together. In a separate test set, we counted how many rare co-activations occurred with each input, and grouped the inputs based on whether its class…

machine learningkoneoppiminenerror detectionvirheetfault toleranceneuroverkotneural networksconcept driftluotettavuusdependability
researchProduct

Emotion Based Music Recommendation System

2020

Nowadays, music platforms provide easy access to large amounts of music. They are working continuously to improve music organization and search management thereby addressing the problem of choice and simplify exploring new music pieces. Recommendation systems gain more and more popularity and help people to select appropriate music for all occasions. However, there is still a gap in personalization and emotions driven recommendations. Music has a great influence on humans and is widely used for relaxing, mood regulation, destruction from stress and diseases, to maintain mental and physical work. There is a wide range of clinical settings and practices in music therapy for wellbeing support.…

machine learningkoneoppiminenrecommendation systemtunteetlcsh:TK5101-6720musiikkisuosittelujärjestelmätsuosituksettekoälyartificial intelligencemusic curationlcsh:Telecommunication
researchProduct

Matemātiskās un statistiskās datu analīzes metodes cukura diabēta pētījumos

2020

Šajā darbā aprakstītas matemātiskās un statistiskās datu analīzes metodes, kas pielietotas cukura diabēta pētījumos. Tā kā cukura diabēts ir viena no izplatītākajām slimībām pasaulē, un tās slimnieku skaits ar katru gadu palielinās, ir nepieciešams izstrādāt tādus matemātiskos modeļus, kas prognozētu personas iespējamību saslimt, tādejādi spējot laicīgi veikt preventīvus pasākumus. Darbā apskatītās metodes tika implementētas brīvpiekļuves programmā R, un ar to palīdzību tika analizēti Latvijā ievākti dati, kā arī mašīnmācīšanās algoritmu izstrādē pasaulē populārā iebūvēto datu kopa PimaIndiansDatabase.

machine learningmatemātiskie modeļidiabetes mellitusMatemātikacukura diabētsmašīnmācīšanās
researchProduct