This study introduces a dataset for drug-bacteria associations (DBA) that affects humans. Our contribution extends beyond merely curating the association matrix; we also conduct genomic similarity computations for all known bacteria impacting humans and assess structural similarities for all antibiotic drugs. This is a first of a kind dataset serving the fundamental goal of streamlining antibiotic repurposing for bacterial infections. we conducted thorough benchmarking on this dataset using state-of-the-art drug repurposing models. Furthermore, we applied this dataset to three case studies involving drug-resistant bacterial infections. Our results indicate that the dataset coupled with the algorithms are indeed effective for repurposing drugs with the potential for treating drug resistant bacterial infections.