Low-frequency distributed acoustic sensing (LFDAS) signal has been used to detect fracture hits at offset monitor wells during hydraulic fracturing operations. Typically, fracture hits are manually identified, which can be subjective and inefficient. We implemented machine learning-based models using supervised learning techniques in order to identify fracture zones, which demonstrate a high probability of fracture hits automatically. Several features are designed and calculated from LFDAS data to highlight fracture-hit characterizations. A simple neural network model is trained to fit the manually picked fracture hits. The fracture-hit probability, as predicted by the model, agrees well with the manual picks in training, validation, and test data sets. The algorithm was used in a case study of an unconventional reservoir. The results indicate that smaller cluster spacing design creates denser fractures.