How To Help Students Choose University Classes: Japanese Case Study

How To Help Students Choose University Classes: Japanese Case Study
Summary: Given the importance of appropriate course selection for students, it’s important for the course recommendation system to not be static, but instead to be designed for proper interactivity and intelligibility. Here is a quick overview of a Japanese university that tried tackling this issue.

Based On The CourseQ Study (2021)

Many of us have been first-year university students at some point in life—excited but totally lost. From being lost on a physical campus to being lost in a university Learning Management System, starting university can be intimidating. Course catalogs are no exception: the variety of options is often fascinating and overwhelming at the same time. Japanese Instructional Designers empathized with students' pain and recently developed an interactive course recommendation system, called CourseQ, that allows students to explore courses by using a novel visual interface to improve transparency and user satisfaction with course recommendations.

Course Recommendation System For Students

Given the importance of appropriate course selection for students, it’s integral for the recommendation system to be designed for good interactivity and intelligibility. By involving students in the recommendation process, the system can capture user preferences interactively. The authors of the CourseQ study point out that while algorithms aimed at finding movie or music recommendations filter purely based on potential interest, course recommendation is a more nuanced task. Students surely want to take interesting courses, but they also need their courses to check a number of other boxes. The course term, the time slots offered, any degree or prerequisite requirements, what courses previous students in their degree program have taken—all of these factors can have a meaningful impact on what courses a student will choose to take and how well they will do in those courses.

Using all of these different factors, and more, when searching for courses can enrich the course recommendation experience for all types of users. As the authors suggest, students who have not yet picked a degree trajectory or just have general interests can browse through available courses based on scheduled class time or term, or by keywords relating to their interests. Students who have a clear learning path in mind may find it more useful to search based on more concrete factors such as prerequisite courses or what peers in previous years have taken. A more robust recommendation system such as CourseQ, as opposed to a more traditional algorithm, allows for many different ways to find relevant courses. Here is an example of the interface:

Students who know exactly what they are looking for can use the search bar in the top right corner to find their course. For students who need help finding course recommendations, functions such as filters and keyword input are found at the top of the interface. The keywords selected by the student are used as seeds for recommendations. Students can further filter course recommendations based on their needs (e.g., degree program requirements, the length of the course, schedule, etc.). Moreover, CourseQ extracts historical enrollment data from within the system, meaning students can find courses that have been popular in the past within different departments. This can help students assess which courses will be most useful to them while building their learning path.

Using keywords associated with a student's interests and course data from the school, the system recommends courses and creates data points (or nodes) within the interface. This is done by calculating the Euclidean distance between the vectors of "student interest" and "courses," and arranging recommended courses based on their similarity to the student’s interests. The nodes created by this algorithm are conveniently color-coded to aid the visualization of the data. The topics used to find the courses are represented by different colors, with a key, defining them, found in the upper right corner. Students can then click on the nodes of recommended courses to learn more about them, including the course description, instructor, schedule, and location. Students can also "like" a course. Each time a student "likes" a course, this data is added to the algorithm to help calculate further interests and keywords. Students are able to edit their "like" lists, allowing them the chance to give real-time feedback within the system.

Allowing students to visualize the courses that will work best for them is a big step forward in course recommendation. While CourseQ may be complex, it can save students the major headaches that come with accidentally choosing courses that don't fit with their preferred learning path. As the researchers point out, "the cost of making an inappropriate decision is much higher for students than investing two hours in a movie they don't like or listening to a song they aren’t interested in." Choosing courses can have a huge impact on a student's life. Maybe it's time for a more comprehensive course recommendation system.


Ma, B., M. Lu, Y. Taniguchi, et al. 2021. "CourseQ: the impact of visual and interactive course recommendation in university environments." Research and Practice in Technology Enhanced Learning 16 (18):

Image Credits:

Screenshot of the CourseQ Interface. From Ma, B., M. Lu, Y. Taniguchi, et al, "CourseQ: the impact of visual and interactive course recommendation in university environments." 2021. RPTEL 16 (18): Provided by the author under the Creative Commons License.