In certain platforms, such as Google Docs, documents are adapted for specific mobile device types, and the installation of their applications is required. Although the documents can be accessed via Web browsers, their correctness is not guaranteed. In content selection, the document is adapted into various versions, from which the optimal one, based on a quality of experience (QoE) criterion, is delivered. Existing works evaluate the QoE of each content version using the user's preferences and context parameters, such as device resolution and network bitrate. They ask the user to weight the context parameters, and then combine them using the simpleadditive-weighting (SAW) method. However, not all users are familiar with the context parameters, and cannot understand their relationship with the requested content. Besides, not all parameters are compensatory to be summed up. In this paper, we propose a TOPSIS-based QoE model to address the two aforementioned drawbacks. We use the context parameters to define high-level functions understandable by all users, and combine them using the TOPSIS method. Experimental results show the convenience of our QoE model and its reliability over the SAW-based method, as well as the weighting-product (WP) method, which is used as an alternative to the SAW one.