1,720,980 research outputs found
Likelihood Distortion and Bayesian Local Robustness
No full textRobust Bayesian analysis has been mainly devoted to detecting and measuring robustness w.r.t. the prior distribution. Many contributions in the literature aim to define suitable classes of priors which allow the computation of variations of quantities of interest while the prior changes within those classes. The literature has devoted much less attention to the robustness of Bayesian methods w.r.t. the likelihood function due to mathematical and computational complexity, and because it is often arguably considered a more objective choice compared to the prior. In this contribution, we propose a new approach to Bayesian local robustness, mainly focusing on robustness w.r.t. the likelihood function. Successively, we extend it to account for robustness w.r.t. the prior, as well as the prior and the likelihood jointly. This approach is based on the notion of distortion function introduced in the literature on risk theory. The novel robustness measure is a local sensitivity measure that turns out to be very tractable and easy to compute for several classes of distortion functions. Asymptotic properties are derived, and numerical experiments illustrate the theory and its applicability for modelling purposes
Occupational diseases risk prediction by genetic optimization. Towards a non-exclusive classification approach
No full textThis paper deals with the health risk prediction problem in workplaces through computational intelligence techniques. The available dataset has been collected from the Italian Local Health Authority (ASL) as part of the Surveillance National System. The main aim of this work is the design of a software application that can be used by occupational physicians in monitoring workers, performing a risk assessment of contracting some particular occupational diseases. The proposed algorithms, based on clustering techniques, includes a genetic optimization in order to automatically determine the weights of the adopted distance measure between patterns and the number of clusters for the final classifier’s synthesis. In particular, we propose a novel approach, consisting in defining the overall classifier as an ensemble of class-specific ones, each trained to recognize patterns of risk conditions characterizing a single pathology. First results are encouraging and suggest interesting research tasks for further system development.This paper deals with the health risk prediction problem in workplaces through computational intelligence techniques. The available dataset has been collected from the Italian Local Health Authority (ASL) as part of the Surveillance National System. The main aim of this work is the design of a software application that can be used by occupational physicians in monitoring workers, performing a risk assessment of contracting some particular occupational diseases. The proposed algorithms, based on clustering techniques, includes a genetic optimization in order to automatically determine the weights of the adopted distance measure between patterns and the number of clusters for the final classifier’s synthesis. In particular, we propose a novel approach, consisting in defining the overall classifier as an ensemble of class-specific ones, each trained to recognize patterns of risk conditions characterizing a single pathology. First results are encouraging and suggest interesting research tasks for further system development
Predizione del rischio di malattie lavoro-correlate attraverso analisi di clustering e ottimizzazione genetica
Full text linkIl lavoro di questa tesi è stato condotto attraverso le tecniche di intelligenza computazionale per uno studio sulla predizione dei rischi per la salute nei posti di lavoro. Il dataset disponibile è stato popolato da parte delle Aziende Sanitarie Locali (ASL) nell’ambito di un programma per la realizzazione del Sistema Nazionale di Sorveglianza per le malattie professionali e gli infortuni mortali. Lo scopo principale di questo lavoro è la progettazione di un’applicazione software capace di evidenziare situazioni di maggior criticità per la manifestazione di Malattie Professionali che possa essere usata agevolmente da parte dei medici del lavoro come strumento di supporto nella loro attività di prevenzione e sorveglianza della salute dei lavoratori. Gli algoritmi proposti, utilizzano tecniche di clustering e l’ottimizzazione genetica per determinare in maniera automatica sia i pesi delle caratteristiche prese in considerazione nel calcolo della distanza interindividuale che il numero di cluster per la sintesi del classificatore finale. In particolare, si propone un nuovo approccio che consiste nel definire il classificatore generale come un insieme di classificatori specifici per ciascuna classe di patologia, ciascuno addestrato a riconoscere le condizioni di rischio che caratterizzano una singola patologia. I primi risultati sono incoraggianti e suggeriscono interessanti temi di ricerca per un ulteriore sviluppo del sistema.The study of this research deals with the health risk prediction problem in workplaces through computational intelligence techniques. The available dataset has been collected from the Italian Local Health Authority (ASL) as part of the Surveillance National System. The main aim of this work is the design of a software application that can be used by occupational physicians in monitoring workers, performing a risk assessment of contracting some particular occupational diseases. The proposed algorithms, based on clustering techniques, includes a genetic optimization in order to automatically determine the weights of the adopted distance measure between patterns and the number of clusters for the final classifier’s synthesis. In particular, we propose a novel approach, consisting in defining the overall classifier as an ensemble of class-specific ones, each trained to recognize patterns of risk conditions characterizing a single pathology. First results are encouraging and suggest interesting research tasks for further system development
Occupational diseases risk prediction by cluster analysis and genetic optimization
No full textThis paper faces the health risk prediction problem in workplaces through computational intelligence techniques applied to a set of data collected from the Italian national system of epidemiological surveillance. The goal is to create a tool that can be used by occupational physicians in monitoring visits, as it performs a risk assessment for workers of contracting some particular occupational diseases. The proposed algorithm, based on a clustering technique is applied to a database containing data on occupational diseases collected by the Local Health Authority (ASL) as part of the Surveillance National System. A genetic algorithm is in charge to optimize the classification model. First results are encouraging and suggest interesting research tasks for further systems' development
Supervised machine learning techniques and genetic optimization for occupational diseases risk prediction
No full textWorkers healthcare gained a lot of attention recently as many countries are increasingly concerning about welfare. This paper faces the problem of predicting occupational disease risks by means of computational intelligence and pattern recognition techniques. Specifically, three different machine learning approaches are compared: the first one is based on the k-means algorithm, in charge to determine a set of meaningful labelled clusters as the final model. The latter two are based on fully supervised techniques, namely Support Vector Machines and K-Nearest Neighbours. Real data regarding both the worker and the workplace by mixing numerical and categorical attributes have been used for testing. The three approaches are automatically tuned by means of genetic algorithms in order to simultaneously find the optimal hyperparameters for the classification systems and the optimal ad-hoc dissimilarity measure weights in order to maximize the classification performances. Computational results show that the three approaches are rather comparable in terms of performances, but a clustering-based approach allows a deeper knowledge discovery phase, helpful for further risk assessment and forecasting
Bayesian size-and-shape regression modelling
Full text linkBuilding on Dryden et al. (2021), this note presents the Bayesian estimation
of a regression model for size-and-shape response variables with Gaussian
landmarks. Our proposal fits into the framework of Bayesian latent variable
models and allows a highly flexible modelling framework
Censoring heavy-tail count distributions for parameter estimation with an application to stable distributions
No full textA new approach based on censoring and moment criterion is introduced for parameter estimation of count distributions when the probability generating function is available even though a closed form of the probability mass function and/or finite moments do not exist
Tropospheric ozone column retrieval from OMI data by means of neural networks: a validation exercise with ozone soundings over Europe
No full text
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
