METHOD AND DEVICE FOR OPTIMIZING INFORMATION DIFFUSION BETWEEN COMMUNITIES LINKED BY INTERACTION SIMILARITIES
A device (OD), intended for optimizing diffusion of information into communities of social network(s) represented by a graph, comprises: —an analysing means (ICE) for analysing this information to determine at least one meaningful concept defining it, —a building means (IRE) for building an ordered list of communities according to numbers of determined meaningful concepts matching respectively concepts associated to the communities of the list, —a filtering means (FM) for defining a group of target communities as entry points for initiating the diffusion of the information, by filtering this built ordered list, —a simulation means (DGCIS) for simulating the information diffusion on the graph (G), and —a processing means (PM) for allowing the choice of at least one target community into the defined group from at least the simulated information diffusion, and then for diffusing the information in each chosen target community.
1 . Method for optimizing diffusion of information, by means of a diffusion service, into communities of social network(s), represented by a graph based on measures of similarities between interactions of community pairs,
during which one analyses said information to determine at least one meaningful concept defining it,
during which one builds an ordered list of communities according to numbers of determined meaningful concepts matching respectively concepts associated to the communities of said list,
during which one defines a group of target communities as entry points for initiating the diffusion of said information, by filtering said built ordered list,
during which one simulates said information diffusion on said graph, and
during which one chooses at least one target community into said defined group from at least said simulated information diffusion, and then one diffuses said information in each chosen target community by means of said diffusion service.
2 . Method according to claim 1 , wherein during which one builds an ordered list of communities according to numbers of determined meaningful concepts matching respectively concepts associated to the communities of said list said communities of said ordered list are extracted from their interactions.
3 . Method according to claim 1 , wherein during which one simulates said information diffusion on said graph one simulates said information diffusion on said graph to obtain a virtual graph, then one computes reachable communities from said virtual graph, and during which one chooses at least one target community into said defined group from at least said simulated information diffusion, and then one diffuses said information in each chosen target community by means of said diffusion service one chooses each target community from said computed reachable communities.
4 . Method according to claim 3 , wherein it further comprises during which one collects service information from said communities to update said graph, and then one compares said updated graph with said virtual graph to compute a final return on investment.
5 . Method according to claim 4 , wherein one determines an audience level from said collected service information, and, if said determined audience level is not satisfactory, one adapts said information to be diffused.
6 . Method according to claim 1 , wherein during which one chooses at least one target community into said defined group from at least said simulated information diffusion, and then one diffuses said information in each chosen target community by means of said diffusion service before choosing each target one evaluates costs of said simulated information diffusion according to a grid of prices, calculated according to targets to be reached, and then one chooses each target community from said cost evaluation.
7 . Method according to claim 6 , wherein during which one chooses at least one target community into said defined group from at least said simulated information diffusion, and then one diffuses said information in each chosen target community by means of said diffusion service one chooses each target community from said cost evaluation to optimize an expected return on investment.
8 . Device for optimizing diffusion of information, by means of a diffusion service, into communities of social network(s), represented by a graph based on measures of similarities between interactions of community pairs, said device comprising:
an analysing means arranged for analysing said information to determine at least one meaningful concept defining it,
a building means arranged for building an ordered list of communities according to numbers of determined meaningful concepts matching respectively concepts associated to the communities of said list,
a filtering means arranged for defining a group of target communities as entry points for initiating the diffusion of said information, by filtering said built ordered list,
a simulation means arranged for simulating said information diffusion on said graph, and
a processing means arranged for allowing the choice of at least one target community into said defined group from at least said simulated information diffusion, and then for diffusing said information in each chosen target community by means of said diffusion service.
9 . Device according to claim 8 , wherein said simulation means is arranged for simulating said information diffusion on said graph to obtain a virtual graph.
10 . Device according to claim 9 , wherein said processing means is arranged for computing reachable communities from said virtual graph.
11 . Device according to claim 9 , wherein said processing means is arranged for comparing said virtual graph with an updated graph resulting from an update of said graph, following upon said information diffusion, and for computing a final return on investment from said comparison.
12 . Device according to claim 8 , wherein said processing means is arranged for evaluating costs of said simulated information diffusion according to a grid of prices, calculated according to targets to be reached.
13 . Server, wherein it comprises a device according to claim 8 .