Parietography

Parietograph is a method of analyzing a pedigree graphically, placing patients in the form of separate points, stratified by various markers - causes of death, age of onset of the disease, etc. On the graph, each person is represented by a separate point, and these points are connected by lines of varying lengths proportional to the number degrees of kinship between members of these families, and within families (joint members) connections are made by polylines of 2, 3, 4 degrees of kinship. The lines on the graph show a series of generations, usually from great-great-great-grandfather to great-grandchildren and vice versa (Fig.).

Partitus (parthitus, Latin “broken, dissected”) or parthetis graph is a metric private way of depicting genealogical diagrams in the form of a “branched” shaded coordinate diagram of two-dimensional space, reflecting the quantitative and qualitative specificity of variable segments of family ties of family group members in Ameyo-Nakagawa codons and Retgie.

Parietography is a data analysis method that is based on visualization and interpretation of paired comparisons. This method was developed in the 1970s by French psychologist Paul Verlaine and his colleagues at the University of Paris-Sorbonne.

The parietographic technique is based on the use of two tables: a table containing paired comparisons, and a table presenting the results of the analysis of paired comparisons. The first table compares pairs of objects, and the second table presents the results of the analysis of these pairs.

Pairwise comparisons can be represented as a matrix, where each cell corresponds to a pair of objects, and the values ​​in the cell indicate which object is preferred over the other. For example, if a cell says “A is greater,” this means that the first object in the pair is more preferred. If a cell says “Equal to,” it means that both objects in the pair are equal.

Paired comparison analysis involves several steps:

1. Normalization: To make paired comparisons comparable, it is necessary to normalize them to a common measurement scale.
2. Correlation: Calculate the correlation coefficient between pairs of objects.
3. Factor analysis: identifying the underlying factors that explain most of the variance in pairwise comparisons.
4. Cluster analysis: dividing objects into groups based on the similarity of pairwise comparisons.