ClassRank is an application that uses collaborative filtering to help personalize class recommendations. Collaborative filtering is a machine learning algorithm that recognizes users like you. After finding those similar users, it uses a weighted average algorithm. The results of this algorithm are then used to select recommendations. Similar users can be found in a variety of ways, generally the algorithm uses a variant of the distance formula.
Once the system finds personalized ratings, it can present them to the user in a variety of ways. It can use them to recommend courses, professors, or schedules. This provides an incentive for students to rate professors. As the students provide truthful ratings, the analytics they receive becomes more accurate. As more students join, the accuracy of the system increases because there are more users to compare to.
The system is then able to solve a common problem: that students are different and what is easy for me may not be easy for you. Our team brings experience in machine learning, which encompasses collaborative filtering. We also bring experience in web development and front-end design. This means that we have the technical skills to create the complex analytic engine and the web design skills to make the application useful.
We believe that we can create a successful application that collects, analyzes, and presents user data. We think that we can do this while developing with best practices. The end result should be a useful, extensible application with long-term applicability.