| CPC G06Q 30/0202 (2013.01) [G06Q 30/0201 (2013.01); G06Q 30/0282 (2013.01); G06Q 40/06 (2013.01)] | 18 Claims |
|
1. A computer-implemented method, comprising:
based on constructing a latent network of a plurality of stakeholders comprising buyers, sellers, museums and art institutions that drive values in an art market, identifying a plurality of influencers within the plurality of stakeholders in the art market and measuring impact factors of the plurality of influencers, wherein the plurality of influencers comprises a plurality of producers and a plurality of institutions;
gathering data representing a plurality of events defining relationships between the plurality of producers and the plurality of institutions, each event comprising information defining at least a time of event and a type of event;
determining a trajectory for each producer of the plurality of producers, the trajectory for a particular producer including a group of events of the plurality of events that involve the particular producer;
based on the trajectory for each producer of the plurality of producers, generating a dynamic graph specifying the relationships between the plurality of producers and the plurality of institutions for a hyperparameter time period;
generating a projection graph from the dynamic graph such that the projection graph specifies only the relationships between institutions of the plurality of institutions;
generating a rating for each institution of the plurality of institutions specified in the projection graph, comprising:
identifying, on the projection graph, a target node of the plurality of institutions with a propagated rating value;
identifying connecting institutions that are connected to the target node on the projection graph;
determining the rating for each connecting institution based on how many connections each connecting institution has to other institutions on the projection graph; and
recommending a category of producers based on the determined rating for each connecting institution;
determining a trajectory score for each producer, the trajectory score for the particular producer being a summation of the rating of each institution that has a relationship with the particular producer over the hyperparameter time period;
determining a proportional demand value, wherein the proportional demand value is a total trajectory score divided by a summation of the trajectory scores for each producer in the plurality of producers over the hyperparameter time period;
determining a performance score for each institution, the performance score for a particular institution being a summation of a delta trajectory score of each producer that has a relationship with the particular institution over the hyperparameter time period; and
outputting a recommendation to invest in an institution in the art market having at least one of a high proportional demand value or a high performance score.
|