1,721,042 research outputs found
Proceedings of the 16-th International Conference on Statistical Analysis of Textual Data
No full text
Clusering of financial time series: a bibliometric analysis
No full textClustering is nowadays widely applied in finance, for solving portfolio selection and risk management problems. In this paper, we propose a review related to both state-of-the-art and of the recent developments of this approach. We adopt a bibliometric analysis, mapping the main issues discussed by scholars in the last 30 years with a network-based technique known as thematic analysis
Conjoint Analysis based methodologies for the ex-ante evaluation of regulatory impact
Full text linkThe activity of evaluation of Public Intervention, or Regulation activity, is actually considered from public administration as a strategic element of political and administrative action. This gives rise to the development of several methods for the ex-ante evaluation of the effects of normative regulations, both on citizens and enterprise activities and on organization and operation of Public Administrations. However, the proposed methodologies not taking into account the complexity and the multidimensionality of the phenomenon, often offer a partial and qualitative point of view. Here we propose several statistical methods based on the classical Conjoint Analysis model. Our aim is to measure and evaluate the sustainability and the expected benefits of regulation respect to different designed alternatives. Mainly, we propose to apply a strategy that - integrating the Conjoint Analysis with graphical factorial representations - allows getting several purposes such as to synthesize individual judgments and to underline the different evaluation preference structures expressed by several groups of judges. The developed methodologies will be applied on real data
Dimensionality Reduction of Unstructured and Network Data for Stance Detection
Full text linkThe idea behind this work stems from the participation in some shared tasks concerning stance detection in NLP conferences. In these competitions, participants tried to develop the best stance prediction system for 'favor', 'against', and 'none' categories on selected topics, according to messages and relationships among users of a social networking site. Thus, the data available consisted of textual and network data. The teams we collaborated with used dimensionality reduction methods for network data, through a Multidimensional Scaling. On the other hand, the approach towards textual data involved different methods of feature extraction, without paying particular attention to dimensionality reduction for unstructured data. In this paper we show the empirical results of a two-step strategy to obtain lower-dimensional textual data relying on text mining techniques and principal component analysis. The results show levels of accuracy comparable to classical feature extraction techniques and to the best task models, despite using a much smaller number of predictors
Proceeding of the 16th International Conference on Statistical Analysis of Textual Data (JADT22)
No full text
- …
