2 Aggregate linguistic analysis: a cooking recipe

Defining the list of features

Usually there is a large number of linguistic features that may be of potential interest to the study, however, in order to conduct a feasible analysis one will have to limit the features, while simultaneously aiming to integrate as many as possible in order to ensure comprehensiveness and reduce the degree of subjectivity.

Creating a feature matrix with N (number of objects) x p (number of features) dimensions

N could for instance refer to the number of dialects under investigation, while p could represent the various features based on which the analysis of the dialects is conducted.

Aggregate

This can be regarded as the main step of an aggregate data analysis. The previously defined features are assembled into more wieldy superordinate groups by means of some sort of dimension reduction technique.

Analyse, visualize and interpret

Created with the Personal Edition of HelpNDoc: Free Web Help generator