A Network-Based Approach to Modeling and Predicting Product Coconsideration Relations
Table 1
Constructing explanatory dyadic attributes.
Configuration
Statistic
Dyadic effects
(a) Binary product attributes
Sum variable
Attribute baseline effect
Matching variable
Homophily effect
(b) Categorical product attributes
Matching variable
Homophily effect
(c) Continuous product attributes (standardized)
Sum variable
Attribute baseline effect
Difference variable
Homophily effect
(d) Non- product related attributes
Distance variable
Homophily effect
(i) represents the indicator function; (ii) represents the absolute-value norm on the 1-dimension space; (iii) represents the -norm on the -dimension Euclidian space.