Data mining is an increasing area of interest where the collection of large amount of data is characterized by heterogeneous information with respect to origin and content; thus, a high degree of specialization is required for a correct analysis. In this paper, we limit ourselves to consider opinions that are expressed as ordered preferences and may be delivered as rating or ranking evaluations. Such situations are different and deserve careful considerations. In both cases, we discuss the framework of CUB models introduced to analyse the ordinal responses by which people express their opinions. Specifically, the approach may be inserted as a useful routine in data mining area for improving the study of essential features supported by empirical evidence. © Springer-Verlag Berlin Heidelberg 2013.
A model-based approach for qualitative assessment in opinion mining / Iannario, Maria; Piccolo, Domenico. - (2013), pp. 113-120. (Intervento presentato al convegno Joint Meetings on Classification and Data Analysis Group of the Italian Statistical Society, CLADAG 2010 tenutosi a Firenze, ita nel 2010) [10.1007/978-3-642-28894-4-14].
A model-based approach for qualitative assessment in opinion mining
IANNARIO, MARIA;PICCOLO, DOMENICO
2013
Abstract
Data mining is an increasing area of interest where the collection of large amount of data is characterized by heterogeneous information with respect to origin and content; thus, a high degree of specialization is required for a correct analysis. In this paper, we limit ourselves to consider opinions that are expressed as ordered preferences and may be delivered as rating or ranking evaluations. Such situations are different and deserve careful considerations. In both cases, we discuss the framework of CUB models introduced to analyse the ordinal responses by which people express their opinions. Specifically, the approach may be inserted as a useful routine in data mining area for improving the study of essential features supported by empirical evidence. © Springer-Verlag Berlin Heidelberg 2013.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
