Online media, such as Yelp, TripAdvisor and Amazon, are full of opinionated information that can easily and significantly influence
a large number of customers' decisions.
Due to the ``word-of-mouth'' effect,
dishonest businesses have adopted unethical or even illegal marketing strategies by paying
spammers to post fake reviews (opinion spams) to promote or demote the targets businesses and products, leading to trustworthiness issues
of the online contents.
To address the issue, trustworthy (defined by AIR
="Accurate, Reliable and Interpretable") fraud detections is required (sketched in
We've adopted propagations over networks
Spam detectors are also constantly under attack of adversarial spammers and thus maintaining proactive detectors
Since humans (model developers and users) are in the detection loop, detection reliability and interpretability is desired.
We propose model debugging and interpretation to deliver these desiderata