deep user analysis

How does santA.Inside understand and analyze users?
Whenever users respond to new items, santA.Inside updates the user model by the
responses of other users as well. This enables more accurate analyses with less data.

  • 1. Deep User Analysis Networks

    his network fits a user model vector that captures the user-item response.
    The whole process can be performed in real-time with even better accuracy than
    State-Of-The-Art models using our network RUNs.

    Deep User Analysis Networks architecture
  • 2. RUNs (State-of-the-Art Real-time Modeling Network)

    User models should be updated in real time since users respond to items every second.
    Also, user models should be predicted by a small number of responses
    because there are many new users. RUNs is the network to achieve both sides.

    RUNs (State-of-the-Art Real-time Modeling Network) architecture
Read the paper