In this composition, the author proposes a system for relating homogeneous consumer groups grounded on qualitative data. The problem is that when probing the end- stoner request, information is frequently presented not in quantitative but in qualitative form. The arbitrary variables with which fine statistics deal are generally assumed to be numeric. thus, among experimenters there's an opinion that achieving at least an interval position of dimension is always desirable, since it expands the experimenter capabilities, giving him grounds to use data fine and statistical analysis traditional styles. Sociologists, on the other hand, emphasize the qualitative data enormous part in the repliers' study. The presented methodology is grounded on cluster analysis, differs from the applied request segmentation styles in that it uses cluster analysis algorithms developed concerning qualitative pointers, and involves a propinquity measure use that allows one to determine the natural weights between clustering variables. Also, the fashion provides for the optimal partition determination grounded on the changes' graph in the average internal communication, depending on the named clusters' number. The optimal among the partitions set is considered to be a partition in which the average internal connection increases sprucely in comparison with the former partition. handed that the clusters' number in each posterior partition in comparison with the former bone is lesser by one. therefore, the methodology allows relating the being request structure.
Author(s) Details:
G. S. Gabidinova,
Naberezhnye Chelny Institute (Branch) of Kazan Federal University, 68/19, Mira prospect, Naberezhnye Chelny, Republic of Tatarstan, 423812, Russia.
Please see the link here: https://stm.bookpi.org/CABEF-V5/article/view/8456
No comments:
Post a Comment