Zum Inhalt springen
Technische Universität München

Practical Course in Winter Semester 2025-26:

Hands-on Recommender Systems

Ashmi Banerjee

News

  • [10 June 2025] Course website published!

Important Information

  • Pre-meeting: No pre-meeting
  • Registration: using the matching system
  • Duration: 13.10.2025 – 06.02.2026 (tentative)
  • ECTS: 10.
  • Capacity: 24
  • Concept:
    • Students in groups of maximum 3 people are required to come up with a potential use-case on Tourism Recommender Systems and implement it over the lecture period of 14 weeks.
    • You can either apply together with your team or build your own team during the initial phase.
      • If you already have a team of your own you can send your CV and a short motivation statement (200-250 words) to ashmi[dot]last_name[at]tum[dot]de (cc: Stefan.Neubig[at]outdooractive.com) with the request to match you in the same group.
      • There is no guarantee that you will be matched with your preferred group-mates by the matching system but this definitely increases your chances.
      • Final team formation will be completed on the first day of the lab course.
    • You are expected to come up with an idea that uses Recommender Systems in Travel and Tourism and implement it
    • Since this course is a collaboration with Outdooractive, there will be the opportunity to use data from Outdooractive. However, it is encouraged that students use a combination of multiple data sources to build their system.
    • In the end they will have to present their results during a final presentation.
    • There will be mandatory milestone presentations where each group is required to present their intermediate progress for group feedback/discussion in 15mins (10 mins presentation + 5 mins Q&A).
    • Final presentations at the end of the semester TBD
  • Information on the Presentations
    • Duration is 25-30 minutes (including Q&A) for final presentation and 10-15 mins (including Q&A) for milestone ones.
    • The talk should be given freely, i.e. not completely read out from a script (in English)
    • You should present slides electronically in any format (e.g. Powerpoint or PDF)
    • You can use the TUM powerpoint template or your own format for the slides
  • Information on the Final Report:
    • Maximum 8 pages (excluding references, appendices etc). in English using the 2-column ACM sigconf template, and not the older one with a single column
    • Recommended Overleaf template: https://www.overleaf.com/gallery/tagged/acm-official
    • It should contain the following mandatory information in the appendix:
  • A table mapping the team-members' full names to their GitHub handles, and listing the contributions of each team member (Team Member | GitHub handle | Contribution).
  • If you're building a UI, it should include relevant screenshots in the document. (But make sure that your images do not occupy the whole document and there is also relevant text describing your project).
  • State your name, affiliation, and email address, use your own keywords, and include a short abstract
  • No postal address or telephone number, no permission block, copyright line or page numbering, no categories, and subject descriptors or general terms
  • References and citations need to be in the correct format (but usage of a LaTex BIB file is optional)
  • Acknowledgments or appendices are optional and not expected
  • You can check here for some best practices for scientific writing.
  • Evaluation criteria:
    • GitHub repository with correct accesses and weekly presentations
    • Performance during presentation(s)
    • Final presentation
    • Final report

Course Description

In this course, will emphasize on the hands-on process of developing Tourism Recommender Systems from inception to production.

Travelers today rely on the Internet for information to plan their trips. However, the explosive amount of available digital information brings the potential challenge of information overload. Tourism Recommender Systems play an effective tool for handling this information overload by helping end users find information of their interest and preference.

In this course, students will work in teams of a maximum of three people, on a hands-on project, giving them the opportunity to gain experience in implementing and evaluating recommender systems using real-world data and tools. Since this course is a collaboration with Outdooractive, there will be the opportunity to use data from Outdooractive. However, it is encouraged that in addition to the provided data, the students use a combination of multiple data sources to build their system.

By the end of the course, students will have a deep understanding of how to design and implement effective recommender systems for the tourism industry, and be able to apply this knowledge to their own projects and work in the field.

Procedure

TBD

Meeting Schedule

  • We plan to hold all lectures and meetings mostly online with some optional in-person meetings!
    EXCEPT FOR THE FINAL PRESENTATION TBD WHICH WILL BE HELD ON-SITE IN GARCHING.
  • Attendance of lectures and meetings is mandatory.

Prerequisites

  • Proficiency in programming using Python (required)
  • Good understanding of version controlling such as Git (required)
  • Basic data analysis skills (required)
  • Understanding of Recommender Systems, Deep Learning (good to have but not necessary)

Contact

To top