1,721,219 research outputs found
Introduction to bibliometrics through bibliometrix
No full textNegli ultimi decenni, il numero di pubblicazioni accademiche sta crescendo ad un tasso esponenziale in tutte le discipline e ciò ostacola il consolidamento della conoscenza e l’individuazione di “evidence-based practices”.
Il tema del seminario è Science Mapping, inteso come quell’insieme di tecniche quantitative multidisciplinari che consentono di realizzare un processo di revisione sistematico, trasparente e riproducibile basato sulla misurazione e la descrizione statistica della scienza, degli scienziati o dell'attività scientifica.
In particolare, durante il seminario, si illustreranno alcuni casi studio attraverso l’impiego del software open-source bibliometrix (Aria & Cuccurullo, 2017). bibliometrix supporta l’intero workflow di science mapping, è programmato in R, e si configura come uno strumento flessibile che può essere rapidamente aggiornato e integrato con altri pacchetti R
A Stabilizing Algorithm for Longitudinal Data Analysis Networks
No full textThis paper provides an alternative approach for longitudinal data analysis through
the use of a set of identified parallel networks.
Longitudinal data can be defined as data resulting from the observation of subjects
(human, animals, organization etc.) over time/space. In this field, we propose a
parallel neural network for the analysis of longitudinal categorical data with constant
row-sum, using a learning algorithm based on a simultaneous latent budget
model (SLBM) [3]. A SLBM is a reduced-rank probability model to decompose a
series of T tables of compositional data observed in different times considering the
constraints across the rows or columns defined on the data. The aim of SLBM is to
approximate the generic matrix P(t) of observed budgets through K latent budgets
obtained as the combination of matrix A(t) of mixing parameters and the matrix B(t)
of latent components satisfying restrictions likewise conditional probabilities.
A problem, well known in literature, is that the Latent Budget Model is not identifiable.
In this case, it means that the learning algorithm produces unidentified
solution for the parallel network. In this paper, we extend the algorithm, called
Stabilizing Algorithm, to identify an unique solution for the parallel network parameters.
The key idea is to use a method based on the structure of the class of
Metropolis algorithm to identify the optimal solution which maximizes the sum
of chi-square distances among the latent budgets.
There are several application fields in which the Parallel Neural Network can be
applied with good results. In particular, this methodology allows to learn a network
that takes into account, simultaneously, the links among the data along the time in
all grounds where this matter is crucial
Systematic Literature Review through science mapping analysis: The Bibliometrix R package
No full textAcademic publications are dramatically growing at fast pace and it is increasingly unfeasible to keep track on all that is being published. Moreover, the emphasis on empirical contributions has resulted in a voluminous and fragmented research streams, and contested field. Nowadays, stand-alone literature reviews are extensively used in various fields for synthetizing findings from previous research, for using effectively the existing base of knowledge and enlarging its boundaries, for providing evidence-based guidelines to practice.
Scholars use various qualitative and quantitative approaches to make sense of earlier findings. Among them, bibliometric analysis is a powerful approach to perform systematic, transparent and reproducible reviews, especially with big volumes of data (big data).
This seminar has two main goals. The first is introducing all the different types of research synthesis. The second is presenting bibliometric analysis for systematic literature reviews. During the seminar, the PhD candidates will perform a bibliometric analysis
BIBLIOMETRIX: An R-tool for comprehensive science mapping analysis
No full textAcademic publications are dramatically growing at fast pace and it is increasingly unfeasible to keep track on all that is being published. Moreover, the emphasis on empirical contributions has resulted in a voluminous and fragmented research streams, and contested field. Nowadays, stand-alone literature reviews are extensively used in various fields for synthetizing findings from previous research, for using effectively the existing base of knowledge and enlarging its boundaries, for providing evidence-based guidelines to practice.
Scholars use various qualitative and quantitative approaches to make sense of earlier findings. Among them, bibliometric analysis is a powerful approach to perform systematic, transparent and reproducible reviews, especially with big volumes of data (big data).
This seminar has two main goals. The first is introducing all the different types of research synthesis. The second is presenting bibliometric analysis for systematic literature reviews. During the seminar, the PhD candidates will perform a bibliometric analysis.
Goals
The seminar will enhance PhD candidates’ skills for bibliometric analysis in the following stages:
• Search: the choice of bibliographic databases; their professional use; the building of a dataset; the cleaning of data.
• Appraisal: the choice of inclusion/exclusion criteria.
• Analysis: the use of R package “bibliometrix”; the descriptive analysis (impact, productivity, trends, bibliometric laws); the analysis of conceptual, intellectual and social structures of knowledge (network analysis; data reduction analysis; thematic evolution).
• Synthesis: data-visualization, with maps, matrices, and networks
Parallel networks for compositional longitudinal data
No full textThis paper provides a new approach to longitudinal data analysis introducing the parallel neural network. Furthermore, is defined a general schema about constraints on the network parameters to deal with compositional data. Finally, we propose an algorithm to solve the identification problem of the parallel network according to the characteristics of longitudinal dat
- …
