Estimating an Activity Driven Hidden Markov Model
This work addresses the challenge of modeling dynamic human behavior for applications in mobility analysis, but it appears incremental as it extends standard HMMs with activity levels without claiming major breakthroughs.
The authors tackled the problem of inferring human mobility on sub-daily time scales from data like mobile phone records by defining a Hidden Markov Model with time-dependent activity levels that drive transitions and emissions, and they showed how to estimate its parameters.
We define a Hidden Markov Model (HMM) in which each hidden state has time-dependent $\textit{activity levels}$ that drive transitions and emissions, and show how to estimate its parameters. Our construction is motivated by the problem of inferring human mobility on sub-daily time scales from, for example, mobile phone records.