On the Use of ‘Ideal Structures’ in Opinion Profile Identification

The concept of ‘ideal structures’, introduced by the author in some preceding studies, is shortly presented along with some considerations, referring to the potential use and significance in multi-modal aggregation of observations, generated through some mechanism, which offers by itself little insight into the potential structure of the resulting data set. An example very much to the point is opinion analysis, based, e.g., on some survey, involving several response dimensions. The aim is to gain knowledge of the structure of the set of opinions with emphasis on the potential variety of coherent opinions. Connections with cluster analysis are indicated and the distinction from this domain is also shown. Some preliminary algorithmic considerations are provided as well.