Product Line (PL) configuration practices have been employed by industries as a mass customization process. However, due to the NP-hard nature of the process, performance concerns start to be an issue when facing large-scale configuration spaces. The aim of my doctoral research is therefore to propose an efficient collaborative-based recommender system that provides accurate and scalable solutions to users. To demonstrate the efficiency of the proposed recommender system, I will conduct series of experiments on real-world extended PLs. In addition, I plan empirically to verify through a user case study the usability of the proposed approach. My expected contribution is to support the adoption of PL configuration practices in industrial scenarios.