# Mining Social Networks Using Heat Diffusion Processes for

Document Sample

```					 國立雲林科技大學
National Yunlin University of Science and Technology

Mining Social Networks Using
Heat Diffusion Processes for
marketing candidates selection
Hao Ma, Haixuan Yang, Michael R. Lyu and Irwin King
CIKM, 2008.

Reported by Wen-Chung Liao, 2009/10/6

1
Intelligent Database Systems Lab
N.Y.U.S.T.
I. M.

Outlines
   Motivation
   Objectives
   Diffusion models
   Marketing candidates selection algorithms
and their complexity
   Empirical analysis
   Conclusions

2                                      Intelligent Database Systems Lab
N.Y.U.S.T.
I. M.

Motivation
   Due to the complexity of social networks, few
models exist to interpret social network
marketing realistically.
   Studies of innovation diffusions, they were
descriptive, rather than predictive
─    they are built at a very coarse level, typically with only
a few global parameters
─   and are not useful for making actual predictions of the
future behavior of the network.

3                                                     Intelligent Database Systems Lab
N.Y.U.S.T.
I. M.

Objectives
   Model social network marketing using Heat
Diffusion Processes.
   Presents three diffusion models, along with
three algorithms for selecting the best

4                                      Intelligent Database Systems Lab
N.Y.U.S.T.
I. M.

HEAT DIFFUSION MODELS
   The process of
people influencing
others is very similar
to the heat diffusion
phenomenon.
   In a social network,
the innovators and
product or innovation
act as heat sources.

5                            Intelligent Database Systems Lab
Diffusion on Undirected Social                                          N.Y.U.S.T.
I. M.

Networks G = (V,E)
G = (V,E)
• V is the vertex set, and V = {v1,
v2, . . . , vn}.
• E is the set of all edges (vi, vj ).
• fi(t) describes the heat at node vi
at time t
• fi(0) initial value.
• f(t) denotes the vector consisting
of fi(t).
• M(i, j, Δt): amount of heat from
node j to node i during a period
Δt

6                                        Intelligent Database Systems Lab
Diffusion on Directed Social Networks                                 N.Y.U.S.T.
I. M.

Diffusion on Directed Social Networks with
Prior Knowledge

7                                  Intelligent Database Systems Lab
Marketing candidates selection
N.Y.U.S.T.
I. M.

O(N(PM+N +N logN))

O(kN(PM +N +d))

O(N(PM +N +kd))

8                                     Intelligent Database Systems Lab
N.Y.U.S.T.
I. M.

EMPIRICAL ANALYSIS
Epinions
• maintains a “trust” list which presents a network of
trust relationships between users,
• product categories, “Kids & Family”
• 75,888 users, and 508,960 edges
• the initial heat vector f(0), choose N/k
• the thermal conductivity value α, set α= 1
• the adoption threshold θ, set θ = 0.6
• t = 0.10, t = 0.15 and t = 0.20, unit???
Scenario:
• 1 to 20 product samples (k =20)
• the marketing candidates?
• performance (measured by the value of coverage) ?

9                                              Intelligent Database Systems Lab
N.Y.U.S.T.
I. M.

EMPIRICAL ANALYSIS

10                    Intelligent Database Systems Lab
N.Y.U.S.T.
I. M.

Conclusion
   Propose a social network marketing
framework which includes three diffusion
models and three marketing candidates
selection algorithms.
   Model social network marketing as
realistically as possible
   Can defend against diffusion of negative
information,
   This framework is scalable.

11                                        Intelligent Database Systems Lab
N.Y.U.S.T.
I. M.

─ Realistic & scalable.

─ Defend against diffusion of negative

information
   Shortage
─ Static social network.

   My opinion:

12                                         Intelligent Database Systems Lab

```
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
 views: 4 posted: 12/11/2011 language: pages: 12
How are you planning on using Docstoc?