Distributing advertisements to distribution target nodes based on a clustering coefficient

• US 7,899,824 B2
• Filed: 11/16/2006
• Issued: 03/01/2011
• Est. Priority Date: 11/17/2005
• Status: Expired due to Fees

• Current Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Original Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Inventors: Miura, Akira, Sasaki, Jun, Akinaga, Yoshikazu, Ihara, Takeshi
• Primary Examiner(s): Vital; Pierre M
• Assistant Examiner(s): Nofal; Christopher P
• Application Number:
  US11/560,595
• Publication Number:
  US 20070136309A1
• Time in Patent Office: 1,566 Days
• Field of Search: None
• US Class Current: 707/737
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0251 Targeted advertisements

Abstract

A cluster extractor extracts a plurality of clusters based on communication records between communication terminals. A clustering coefficient calculator, an average path length calculator, and a degree distribution calculator calculate a clustering coefficient, an average path length in an advertisement distribution target cluster, and a degree distribution, respectively, based on communication records between communication terminals belonging to an advertisement distribution target cluster. An advertising strategy determiner determines an advertising strategy, based on the clustering coefficient and the average path length, and a distribution target determiner determines a distribution target terminal, based on the degree distribution and the advertising strategy. A determined target notifier notifies a communication management device of the distribution target terminal and the advertising strategy and an advertisement distributor distributes an advertisement according to the advertising strategy to the distribution target communication terminal.

Claims

1. An advertisement distributing system comprising:

cluster extracting means for extracting a plurality of clusters, based on communication records between nodes at a plurality of nodes;

clustering coefficient calculating means for, based on communication records between nodes belonging to one cluster extracted by the cluster extracting means, calculating a clustering coefficient indicating a use state of communication in said one cluster;

average path length calculating means for calculating an average path length between nodes belonging to said one cluster, based on the communication records between the nodes belonging to the one cluster;

degree distribution calculating means for calculating a degree distribution indicating node counts for respective correspondent counts of the nodes belonging to said one cluster, based on the communication records between the nodes belonging to the one cluster;

advertising strategy determining means for selecting and determining at least one advertising strategy from a plurality of advertising strategies, based on the clustering coefficient calculated by the clustering coefficient calculating means, the average path length calculated by the average path length calculating means and respective predetermined values, the respective predetermined values being a clustering coefficient and an average path length in a cluster whose node count is maximum out of other clusters extracted by the cluster extracting means;

distribution target determining means for selecting and determining a distribution target node as a target for distribution of an advertisement from the nodes belonging to the one cluster, based on the degree distribution calculated by the degree distribution calculating means and the at least one advertising strategy determined by the advertising strategy determining means; and

distributing means for distributing the advertisement corresponding to the at least one advertising strategy determined by the advertising strategy determining means, to the distribution target node determined by the distribution target determining means.

2. An advertisement distributing system according to claim 1, wherein the advertising strategy determining means selects the at least one advertising strategy including a factor to enable the clustering coefficient calculated by the clustering coefficient calculating means and the average path length calculated by the average path length calculating means to become closer to the respective predetermined values.

3. An advertisement distributing method comprising the following steps executed by a computer:

a cluster extracting step of extracting a plurality of clusters, based on communication records between nodes at a plurality of nodes;

a clustering coefficient calculating step of based on communication records between nodes belonging to one cluster extracted in the cluster extracting step, calculating a clustering coefficient indicating a use state of communication in said one cluster;

an average path length calculating step of calculating an average path length between nodes belonging to said one cluster, based on the communication records between the nodes belonging to the one cluster;

a degree distribution calculating step of calculating a degree distribution indicating node counts for respective correspondent counts of the nodes belonging to said one cluster, based on the communication records between the nodes belonging to the one cluster;

an advertising strategy determining step of selecting and determining at least one advertising strategy from a plurality of advertising strategies, based on the clustering coefficient calculated in the clustering coefficient calculating step, the average path length calculated in the average path length calculating step and respective predetermined values, the respective predetermined values being a clustering coefficient and an average path length in a cluster whose node count is maximum out of other clusters extracted by the cluster extracting step;

a distribution target determining step of selecting and determining a distribution target node as a target for distribution of an advertisement from the nodes belonging to the one cluster, based on the degree distribution calculated in the degree distribution calculating step and the at least one advertising strategy determined in the advertising strategy determining step; and

a distributing step of distributing the advertisement corresponding to the at least one advertising strategy determined in the advertising strategy determining step, to the distribution target node determined in the distribution target determining step.

4. An advertisement distributing system comprising:

a cluster extractor configured to extract a plurality of clusters, based on communication records between nodes at a plurality of nodes;

a clustering coefficient calculator configured to calculate, based on communication records between nodes belonging to one cluster extracted by the cluster extractor, a clustering coefficient indicating a use state of communication in said one cluster;

an average path length calculator configured to calculate an average path length between nodes belonging to said one cluster, based on the communication records between the nodes belonging to the one cluster;

a degree distribution calculator configured to calculate a degree distribution indicating node counts for respective correspondent counts of the nodes belonging to said one cluster, based on the communication records between the nodes belonging to the one cluster;

an advertising strategy determiner configured to select and determine at least one advertising strategy from a plurality of advertising strategies, based on the clustering coefficient calculated by the clustering coefficient calculator, the average path length calculated by the average path length calculator and respective predetermined values, the respective predetermined values being a clustering coefficient and an average path length in a cluster whose node count is maximum out of other clusters extracted by the cluster extractor;

a distribution target determiner configured to select and determine a distribution target node as a target for distribution of an advertisement from the nodes belonging to the one cluster, based on the degree distribution calculated by the degree distribution calculator and the at least one advertising strategy determined by the advertising strategy determiner; and

a distributor configured to distribute the advertisement corresponding to the at least one advertising strategy determined by the advertising strategy determiner, to the distribution target node determined by the distribution target determiner.