The ages of solar-like stars have been at the center of many studies such as exoplanet characterization or Galactic-archeology. While ages are usually computed from stellar evolution models, relations linking ages to other stellar properties, such as rotation and magnetic activity, have been investigated. With the large catalog of 55,232 rotation periods, P rot, and photometric magnetic activity index, S ph from Kepler data, we have the opportunity to look for such magneto-gyro-chronology relations. Stellar ages are obtained with two stellar evolution codes that include treatment of angular momentum evolution, hence using P rot as input in addition to classical atmospheric parameters. We explore two different ways of predicting stellar ages on three subsamples with spectroscopic observations: solar analogs, late-F and G dwarfs, and K dwarfs. We first perform a Bayesian analysis to derive relations between S ph and ages between 1 and 5 Gyr, and other stellar properties. For late-F and G dwarfs, and K dwarfs, the multivariate regression favors the model with P rot and S ph with median differences of 0.1% and 0.2%, respectively. We also apply Machine Learning techniques with a Random Forest algorithm to predict ages up to 14 Gyr with the same set of input parameters. For late-F, G and K dwarfs together, predicted ages are on average within 5.3% of the model ages and improve to 3.1% when including P rot. These are very promising results for a quick age estimation for solar-like stars with photometric observations, especially with current and future space missions.