Lyman alpha emitters (LAEs) in the Epoch of Reionization (EoR) offer valuable probes of both early galaxy evolution and the process of reionization itself; however, the exact evolution of their abundance and the nature of their emission remain open questions. We combine samples of 229 and 349 LAE candidates at z = 5.7 and z = 6.6, respectively, from the SILVERRUSH narrowband survey with deep Low Frequency Array (LOFAR) radio continuum observations in the European Large Area Infrared Space Observatory Survey-North 1 (ELAIS-N1) field to search for radio galaxies in the EoR and study the low-frequency radio properties of z ≳ 5.7 LAE emitters. Our LOFAR observations reach an unprecedented noise level of ~20 μJy beam−1 at 150 MHz, and we detect five candidate LAEs at >5σ significance. Based on detailed spectral energy distribution modelling of independent multi-wavelength observations in the field, we conclude that these sources are likely [OII] emitters at z = 1.47, yielding no reliable z ≳ 5.7 radio galaxy candidates. We examine the 111 z = 5.7 and z = 6.6 LAE candidates from our panchromatic photometry catalogue that are undetected by LOFAR, finding contamination rates of 81–92% for the z = 5.7 and z = 6.6 subset of the LAE candidate samples. This subset of the full sample is biased towards brighter magnitudes and redder near-infrared colours. The contamination rates of the full sample will therefore likely be lower than the reported values. Contamination of these optically bright LAE samples by likely [OII] emitters is lowered significantly through constraints on the near-infrared colours, highlighting the need for infrared observations to robustly identify bright LAEs in narrowband surveys. Finally, the stacking of radio continuum observations for the robust LAE samples yields 2σ upper limits on radio luminosity of 8.2 × 1023 and 8.7 × 1023 W Hz−1 at z = 5.7 and 6.6, respectively, corresponding to limits on their median star-formation rates of <53 and <56 M⊙ yr−1.

Context. Extragalactic radio continuum surveys play an increasingly more important role in galaxy evolution and cosmology studies. While radio galaxies and radio quasars dominate at the bright end, star-forming galaxies (SFGs) and radio-quiet active galactic nuclei (AGNs) are more common at fainter flux densities. Aims. Our aim is to develop a machine-learning classifier that can efficiently and reliably separate AGNs and SFGs in radio continuum surveys. Methods. We performed a supervised classification of SFGs versus AGNs using the light gradient boosting machine (LGBM) on three LOFAR Deep Fields (Lockman Hole, Boötes, and ELAIS-N1), which benefit from a wide range of high-quality multi-wavelength data and classification labels derived from extensive spectral energy distribution (SED) analyses. Results. Our trained model has a precision of 0.92±0.01 and a recall of 0.87±0.02 for SFGs. For AGNs, the model performs slightly worse, with a precision of 0.87±0.02 and a recall of 0.78±0.02. These results demonstrate that our trained model can successfully reproduce the classification labels derived from a detailed SED analysis. The model performance decreases towards higher redshifts, which is mainly due to smaller training sample sizes. To make the classifier more adaptable to other radio galaxy surveys, we also investigate how our classifier performs with a poorer multi-wavelength sampling of the SED. In particular, we find that the far-infrared and radio bands are of great importance. We also find that a higher signal-to-noise ratio in some photometric bands leads to a significant boost in the model performance. In addition to using the 150 MHz radio data, our model can also be used with 1.4 GHz radio data. Converting 1.4 GHz to 150 MHz radio data reduces the performance by ~4% in precision and ~3% in recall.