We show preliminary results of searches for interacting galaxies over an area of about 1000 square degrees in the southern hemisphere, using the second data release of the Southern Photometric Local Universe Survey. Our first goal is to find very nearby interacting galaxies (z < 0.019) with a high fraction of gas in the Fornax cluster and other nearby systems, taking advantage of the large area covered by S-PLUS with images in the Hα band. Our second goal is to obtain a sufficiently representative dataset of interacting galaxies with r < 17.5 mag, using different methods, to feed a supervised training of a deep learning model that is capable of finding new interacting galaxies in S-PLUS automatically. Here we show a sample of interacting galaxies found by the deep learning model under development as a partial result of the project. The accuracy of the model in the test set is \(\approx\)91%.