A A A+ A++

AI Data Wizard – How Many Landslides?

In last year’s AI Challenge, students made use of machine learning/deep learning in AI for weather forecast, where they experienced how to use technology to solve real-life problems. In Master Code 2019, Hong Kong Education City (HKEdCity), Microsoft, Hong Kong Observatory, Hong Kong Meteorological Society and Civil Engineering and Development Department (CEDD) jointly organise the 'AI Data Wizard - How Many Landslides?’ for Secondary school students. Participants will learn how to select suitable data and analyse using machine learning platforms in order to estimate the number of landslides in certain periods.


  1. To empower teachers and secondary school students to make use of machine learning/deep learning in AI
  2. To let students experience how to solve real-life problems using AI
  3. To enhance students’ awareness of the environment and climate change


  • Secondary School Students




  • The Gear-up Workshop on 19 Oct is now open for registration. Please CLICK HERE to enrol.
  • The training data set is now available. Please go to the "Useful Information" Tab to view and download.


Rules & Regulations

  • Each school can form two participating teams maximum, and each team should consist of two to five students. Each student can only participate in one team.
  • The participating teams need to made use of machine learning/deep learning in AI to estimate the total number of landslides in Hong Kong during a few rainstorms in the previous time.
  • Participating teams need to submit the following on or before 12:00 Noon, 25 November:
    • Estimation, AND
    • an introductory PowerPoint (within ten slides) AND
    • an introductory video (within three minutes)
    • to elaborate on how machine learning / deep learning of AI is used in deducing the estimation.
  • Each team can submit once only, no changes can be made after the submission.
  • Submission of any entry indicates that the participant has read and accepted the Rules & Regulations of the competition.
  • The Organisers own the copyright of the awarded entries.
  • The Organisers reserve the right to amend and/or cancel the activities as well as the Rules & Regulations without prior notice.
  • In case of any dispute, the Organisers reserve the right of final decision.


Gear-up Workshop

  • Target: Teachers and Secondary school students who are interested in joining the Challenge
  • Date & Time: 19 Oct (Sat) 9:30 am – 1:00 pm
  • Venue: Room 201-202, 2/F, Hong Kong Education City, Sha Kok Estate, Shatin
  • Content:
    • Introduction of the Challenge
    • Basic concepts on landslides, slope safety with related meteorological knowledge and data
    • Introduction of Machine Learning and hands-on exposure to Azure ML Studio

Students’ enrolment for the activities above shall be completed by schools. 4 participants per school are allowed due to limited seats.



Microsoft          香港天文台          


The judging panel is formed by professionals from coorganisers and follows the below criteria to carry out the judging:

1 Forecast Accuracy 50%
  • The submitted estimation will be compared with the official record. The closer to the official record, the higher mark one will get.
  • The final score of each team will be obtained after normalisation with all other teams' submissions
2 Application of Data analysis and Computational Thinking 20%
  • Select and use suitable data for analysis, adopt mathematical methods to estimate, e.g. average values, maximum and minimum values, etc.
  • Explain the reason of data chosen.
3 Application of Machine Learning 20%
  • Demonstrate the application of machine learning algorithms, e.g. regression, cluster analysis, etc.
  • Explain the use of algorithms in the work
4 Team Collaboration and Communication 10%
  • Clearly show how the team works together by listing the division of labor, etc.



Available Data

Training Data

Please CLICK HERE to download the historical data provided by Hong Kong Observatory Headquarter and Civil Engineering and Development Department (CEDD) for model training.

Please read the instruction file (AI_data_wizard_training_data_Readme.pdf) carefully for a better understanding of the data. The instruction file is also included in the zipped file.


Participants can also use other data from HKO and CEDD website, Newspaper, or any other open data or information from the Internet.


Estimation Data

The criteria and data for estimation will be released by mid-October. 



Learning Kit - Landslides and Slope Safety

Please click here to download


Introductory Workshop - Video Review (28 June)

Part 1


Part 2


Part 3


Awards for 'AI Data Wizard – How Many Landslides?':  

Awards Quantity Prizes
Champion 1 TBC
First Runner-up 1 TBC
Second Runner-up 1 TBC
Merit Awards 5 TBC


  • Each participating student who has completed the competition (submitted all the required works) will be presented a certificate
  • Winners will also be presented relative certificates