Abstract
The measure of expert opinions consistency is an essential component for building ranking models based on user ratings. The article describes a model building method for calculating the measure of opinions consistency for potential experts relative to the current user in the case when it is impossible to apply methods based on their past activities analysis. In the case when the current user does not have common evaluations with potential experts for some set of web resources, the measure of their opinions consistency can be calculated on the basis of social and personal factors. The article analyzes the results of constructing such models using various inductive modeling algorithms and neural network.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have