Data leakage is one of the critical challenges in the area of cloud computing where data sharing is an essential part among multiple entities. This work presents a generic Data Leaker Detection Model (DLDM) that identifies the malicious entity responsible for data leakage. The proposed approach is an integration of cryptography, watermarking and hashing techniques for securing the data. Furthermore, the model detects the venomous user by considering a combination of watermark extraction and probability estimation. The results signify that the time taken to detect the malicious user is 3580 ms when 200 documents of size 20 MB are provided to single/distinct users. The average probability to identify the venomous user is 0:969518 for the load value of 2, which indicates the high probability to identify the guilty agent. The experimental results verify the efficiency of the proposed model.