User susceptibility in the product adoption decisions in a social network
User susceptibility that contributes to sentiment-charged content diffusion in a social network
User susceptibility to online brand-related information while making a brand purchase decision
Node susceptibility (NS) to different types of information diffused in a social network
Quantitative measures
Modeling time to peer adoption as a function of the peer’s treatment status
Defining user susceptibility to be how easy the user adopts the same sentiments diffused by other users
Using 7-point Likert scales (1: strongly disagree to 7: strongly agree)
Defining NS as the probability that quantity of information the user forwards is larger than quantity of information the user receives
Mathematical model
A continuous-time single-failure proportional hazards model:
An influence-susceptibility cynical (ISC) model:
None
A discrete stress-strength interference (DSSI) model based on universal generating function (UGF):
Key input parameters for decisions
() A set of individual attributes of an application user; () baseline hazard; () the number of notifications received by peer
() Item sentiment adopted by user before diffusion; () item sentiment adopted by user after diffusion; () number of followees of user; () number of times sentiment is diffused to user by his followees; () set of item sentiments diffused to user and he adopts the same item; () set of followees who diffuse item sentiment to user and the user adopts the same item sentiment
() Gender; () education; () personal status; () occupation; () age; () income (per month)
() Observation parameter: cumulative quantity of information user received in corresponding time interval; () observation parameter: cumulative quantity of information user forwarded in corresponding time interval
Implementation process in practice
Conducting a randomized experiment to estimate , , , , and
Employing an iterative computation method. The algorithm first initializes susceptibility for all users with 0.5, and the computation process repeats until the values converge
Conducting an online survey on a survey website using a consulting firm panel from Shanghai and Nanjing
Recording the abovementioned observation parameters on corresponding time node and inputting observation parameters to calculate NS
Decision objectives
Identifying which individuals are more susceptible to adopt the product offered
Determining how a user is susceptible to sentiment-charged tweets diffused by others
Determining the extent to which online brand-related information impacted users’ brand attitudes and purchase intentions
Determining three questions: () which users are most susceptible, () which types of information they are most susceptible to, and () when they are most susceptible
