1,720,993 research outputs found
Trajectory-based and Sound-based Medical Data Clustering
No full textChallenges in medicine are often faced as interdisciplinary endeav- ors. In such an interdisciplinary view, sonification of medical data provides an additional sensory dimension to highlight often hard- to-find information and details. Some examples of sonification of medical data include Covid genome mapping [5], auditory repre- sentations of tridimensional objects as the brain [4], enhancement of medical imagery through the use of sound [1]. Here, we focus on kidney filtering-efficiency time-evolution data. We consider the estimated glomerular filtration rate (eGFR), the main indicator of kidney efficiency in diabetic kidney disease patients.1 We propose a technique to sonify the eGFR trajectories with time, frequency, and timbre to distinguish amongst patients (Figure 1). Multiple pitch tra- jectories can be formally investigated with the tools of counterpoint (Figure 2), and computationally analyzed with sound-processing techniques. Patients who present similar patterns of eGFR behavior can be more easily spotted through musical similarities. We use the Fréchet distance, which evaluates the shape similarity between curves [2], to cluster patients with similar eGFR behavior. We thus compare the information gathered through sonification and shape- based analysis. We find the mean curves in each trajectory cluster and we compare them with the characteristics of sonified curves. Clustering methods have also been applied to sound analysis: it is the case of k-means to cluster sound data [3]. The Fréchet-based clustering technique is a development of k-means taking shape into account. Thus, we sketch a sound-based clustering approach for medical data, as an additional tool to find patterns of behavior. This study can foster new research between computer science, medicine, and sound processing
Advances of Spatio-Temporal Clustering For Evaluating The Interaction Between Tourism And Environment
No full textNowadays, there is a growing interest to assess sustainability, which is assuming a key role in various socio-economic areas. In this context, the tourism sector, which is an important factor in terms of economic and social development, has a significant impact on the environment and contributes to climate change. For this reason, there is more and more attention on any forms of sustainable and green tourism and consequently on studies that combine the two underlying domains, that is the tourism growth and the environment safeguard. The aim of this work is to develop a methodological framework for a spatio-temporal clustering model to investigate the in- teraction between tourism and environmental factors in Italy, as well as the pattern recognition in order to reduce the intrinsic heterogeneity of Italian territory. In particular, an innovative boos- trap spatio-temporal clustering is proposed, where the Ward hierarchical clustering algorithm that includes spatial and temporal constraints is included. The space-time component is built as a combination of a spatial distance matrix and a temporal distance matrix based on the con- cept of a space-time metric. This new unsupervised algorithm can be performed for time series that may dier in length. The results obtained by using the proposed procedure are illustrated through a real data set characterized by several tourism and environmental indicators, taken during a five-year period for the administrative regions (NUTS2 level) of Italy. The proposed methodology aims to guide and support the development of regionally tailored environmental policies
Classes of Colors and Timbres: A Clustering Approach
Full text linkSimilarities between different sensory dimensions can be addressed considering common “movements” as causes, and emotional responses as effects. An
imaginary movement toward the “dark” produces “dark sounds” and “dark
colors,” or, toward the “bright,” “brighter colors” and “brighter sounds.”
Following this line of research, we draw upon the confluence of mathematics
and cognition, extending to colors and timbres the gestural similarity conjecture, a development of the mathematical theory of musical gestures. Visual
“gestures” are seen here as paths in the space of colors, compared with paths
in the space of orchestral timbres. We present an approach based on clustering algorithm to evaluate the association between color bands and orchestral
timbres. The analysis is based on 8 indicators which represent and describe
participants’ background and associations to be tested. The indicators include socio-demographic information and color class options from the color
space, to associate with each given timbre class. We clustered responders
into homogeneous groups where the within-group-object dissimilarity is minimized and the between-group-object dissimilarity is maximized. The partitions are obtained with k-modes. While participants’ background does have
an influence in their answers, the overall behaviors confirms the existence
of different space regions for different timbres, supporting our hypothesis of
perceived similarities similarities between color and timbre classes. In fact,
the cluster analysis confirms identifiable blocks. Our pioneering study on a
small dataset may open the way toward a future and deeper comprehension
of complex color-timbre perceived connections
Improving spatial clustering through a weight system on multilevel permanent museum attraction probability
No full textMuseums are extensively distributed all over the Italian territory. In this context, the iden- tification of spatial patterns, referred to specific characteristics of museums evaluated at regional level, can support the enhancement of the cultural and natural heritage as well as the social and economic growth. In the literature, many studies were focused on the visitors’ profile or on the managerial performance and economic efficiency of the muse- ums. However, none of them analysed the effects of the permanent presence of museums and their spatial contiguity by using both spatial machine learning models and statistical models. To this aim an innovative approach, which combines multilevel binary model and spatial clustering, as a machine learning unsupervised technique, is proposed to investigate the pattern recognition of the permanent museums all over the Italian territory and provide relevant information in terms of similarity among the spatial cluster formed. The logit of the museums to remain open all over the year, also with respect to different types of institu- tion (private/public) and a different spatial/geographical constraints are jointly considered. In addition, a weight system is defined in order to introduce a regional measure of muse- ums prevalence with respect to other types of cultural institutions. The ISTAT microdata concerning the Italian survey on museums and cultural entities are considered. The results highlight the great potentiality of this spatial clustering approach in delivering a better understanding of the role of museums as factor of challenge of urban development, provid- ing in the meantime suggestions for tourism providers and museum managers
A statistical model to evaluate the attractiveness of a food and wine event
No full textMultivariate analysis has been applied in order to assess the variables
that influence behaviour and attitudes of visitors of a food and wine event. Stepwise
logistic regression is also used and the most parsimonious set of predictors that are
most effective in predicting the choice of purchase are considere
A WebGIS for Road Accidents Monitoring in an Urban Area
No full textIn the recent years, there is a growing interest in the analysis of spatial distribution of road accidents occurred over a territory, in order to identify the areas which are considered of high risk and improve road safety. In this context, the most advanced information tools, such as Geographical Information System (GIS), allow the collection, management and analysis of large geo-referenced databases concerning qualitative and quantitative information about road accidents. Moreover, a WebGIS, which is a GIS freely usable on the web, makes the data available for all kind of users (policy makers, scientists, analysts) and therefore can be considered a valid support system for planning suitable and effective actions for prevention road accidents. In this paper, the development of a GIS project and the corresponding WebGIS application for road accidents monitoring over an urban area located in the South of Italy, are proposed. Their main features are discussed in order to highlight the potentiality of these tools in using raw statistical and geographical data to produce meaningful information for spatial analysis, mapping and identifying any factors contributing to road accidents
Multi-class random forest model to classify wastewater treatment imbalanced data
Full text linkThe odor emissions generated by treatment plants imply complex environmental and economic issues. The modern instrumental odor monitoring systems, based on an array of several sensors, continuously record the gaseous compounds. However they are characterized by poor selectivity, compromising the possibility to discriminate and identify the emission sources. In this paper, the ability of odor sensors to distinguish between the treatment plant sections generating the gaseous compounds is evaluated on the basis of the random forest classifier, and is also compared to the discriminant analysis performance. Taking into account that a multi- parametric system of sensors can be affected by the presence of a small sample size with imbalanced classes, several strategies for data balancing are proposed and analyzed. The findings show that the random forest classifier is characterized by a better capacity to distinguish the emissions sources with respect to the classical multiple discriminant analysis, in terms of all evaluation metrics. This is also confirmed for different resampling techniques, especially in the over-sampling case. The data concerning measurements from 10 sensors of multi- parametric systems of odor monitoring collected from a company specialized in environmental assistance are considered for this analysis
Multilevel Modeling of Training Needs in Artificial Intelligence
Full text linkNowadays, Artificial Intelligence (AI) is playing a rapidly increasing role in several fields of research and in almost all sectors of real life. However, few studies have assessed the effects of AI applications on training needs. This paper proposes an innovative multilevel modeling in order to investigate Awareness, Attitude and Trust towards AI and their reflections on learning needs. In particular, it is shown how a machine learning variable selection algorithm can support the definition of the optimal subset of all relevant covariates with respect to the outcome variable and improve the multilevel model performance for estimating the probability of educational needs. Thus, starting from a complex web survey to European citizens distributed in eight countries, the estimation of a multilevel binary model, defined on the basis of covariates selected through the Boruta random forest algorithm, is proposed. A discussion on the gender differences of the related estimated multilevel logit models is presented. A sensitivity analysis is also included in order to assess the prediction accuracy of the proposed multilevel logit modeling
Un modello di valutazione della qualità per il servizio di raccolta differenziata
No full textLa misurazione della qualità dei servizi, sia nel settore dei servizi privati che, recentemente, in quello dei servizi pubblici, rappresenta uno strumento fondamentale al fine di favorire e garantire lo sviluppo sostenibile di un territorio. Nel presente lavoro, dopo alcuni cenni sui modelli ServQual e ServPerf, quali strumenti di rilevazione della qualità di un servizio, saranno presentati i risultati di un’indagine campionaria effettuata in un Comune della Provincia di Lecce, riguardante la qualità del servizio di raccolta differenziata dei rifiuti. Successivamente, saranno discussi gli strumenti statistici utilizzati per misurare il livello di soddisfazione degli intervistati per il servizio in esame; infine sarà presentato un WebGis per la gestione dei rifiuti urbani
- …
