The paradigm of data mining-as-a-service in cloud computing environments has garnered significant attention in recent years. In this paradigm, a company (data owner) that lacks the necessary data storage, computational resources, and expertise stores its data in the cloud and outsources its mining tasks to a cloud service provider (server). To safeguard the privacy of the outsourced database and the association rules derived, techniques such as k-anonymity, k-support, and k-privacy have been proposed to perturb the data before it is uploaded to the server. However, these techniques are computationally intensive. If the data owner possesses the resources to employ these techniques, it is often capable of executing association rule mining locally. In this chapter, we explore a scenario in which the data owner encrypts its data and stores it in the cloud. To mine association rules from this encrypted data, the data owner outsources the task to \(n~(\geq 2)\) “semi-honest” servers, which collaborate to perform the association rule mining on the encrypted data in the cloud and return the encrypted association rules to the data owner. We introduce three solutions aimed at ensuring data privacy during association rule mining. These solutions, based on the distributed ElGamal cryptosystem, protect item privacy, transaction privacy, and database privacy, provided that at least one of the n servers remains honest. To minimize the risk of all servers being compromised, the data owner can choose servers from different cloud providers.