eDAADe: An adaptive recommendation system for comparison and analysis of architectural precedents

Abstract

We built a Web-based adaptive recommendation system for students to select and suggest architectural cases when they analyze "Case Study" work within the architectural design studio course, which includes deep comparisons and analyses for meaningful architectural precedents. We applied hybrid recommendation mechanism, which is combining both content-based filtering and collaborative filtering in our suggested model. It not only retains the advantages of a content-based and collaborative filtering approach, but also improves the disadvantages found in both. We expect that the approach would be helpful for students to find relevant precedents more efficient and more precise with their preferences.