Multi-document summarization (MDS) is an automatic process where the essential information is extracted from the multiple input documents.However, it has many limitations such as inaccurate extraction to essential sentences, low coverage, poor coherence among the sentences, and redundancy.So, in this paper, oppositional lion optimization algorithm (OLOA) and deep neural network (DNN) based multi-document summarization is presented. For optimal sentence selection from the pre-processed documents, OLOA algorithm is proposed. Based on the extracted features, sentence score is calculated using DNN. Finally, based on the rank of the sentences, the multi-documents are summarized. Simulation results show that the performance of the proposed summarization outperforms the state of the arts in terms of precision, recall and F-measure.