The Big Data environment has been defined with its main character istics: volume, variety, velocity, and veracity. Since this environment can be exploited for illegal purposes, which highlights the need for digital forensic investigations. However, these mentioned Big data characteristics formed a major challenge for the investigator and the structure of the digital forensics process. While artificial intelligence with its tools has proven an important role in various fields, the so lution came in the form of a proposed framework that incorporates artificial intelligence and automated tools into the digital forensics process. The proposed framework covered all digital forensics phases. Also, we have adopted ensemble learning methods for the digital foren sics analysis phase. Ensemble learning methods are mainly engaged to improve model efficiency. The experimental result came promising and feasible. In fact, in this work, alongside the new proposed frame work, ensemble learning was tried in two cases, the first related to Counterfeit Bank Note, while the second was related to Infiltration from inside incident. The boosting method through the AdaBoost classifier performed best in both cases. This results showed a road for building a comprehensive system.