Sport climbing is a relatively recent sport, and many new climbing routes are continuously built all around the world. Climbers are interested to climb routes which are: challenging, but within their capabilities; enjoyable; suitable for training; and safe. We would like to first understand the preferences of the climbers by monitoring their behaviour, and then to build a recommender system for supporting them in choosing suitable climbing routes.
Modeling a climber’s behavior will be made by using both explicit feedback, e.g., by leveraging data coming from mobile climbing applications (s.a. Vertical Life), and by using implicit feedback, which can come from data acquired by accelerometers placed on the wall or on the body of the climber. By combining these two data sources, we aim at building a climber’s profile and based on such a profile, to generate recommendations.
Aiming at the overall goal of building a climbing routes recommender system, we have now developed machine learning models to predict whether the climber would agree with the “official” difficulty level of the climbing routes. Usually, climbing route difficulty is given by the route setter or a person who initially built the. However, climbers might have different opinions about the grade of a route, hence in disagreement with the route setter. In our application scenario, climbers manually insert their grades for routes that they tried, by using a mobile application (`Vertical-Life’). We use a range of additional data in order to model why they disagree with the setter. We have created a dataset of indoor and outdoor lead climbs, where the official grade and the climbers’ opinions are stored. We have then implemented two ML models to predict the grade difficulty given by the climbers: linear regression and matrix factorization. For the linear regression model, we constructed features of the climber-route interaction which describe how and when the climber deviates from the setter’s grade, and for matrix factorization, we modelled the grade prediction as a special type of rating prediction problem. We have then used singular value decomposition with normalization taking into consideration the route setter’s grade. We show that the models’ predictions are closer to the climbers’ opinions than a baseline model.
In the future, we would like to develop a full system that could be also able to identify climbers in a climbing gym, detect what type of activities they perform, and measure their performance. This system could be used to enrich the climber’s profile and to complete the design of the recommender. Last, but not least, we would like to incorporate the proposed solution into the Vertical-Life app.
[2024]Digital Global Overview Report 2024 Meltwater.pdf
SFScon21 - Ivanova Iustina - Recommendations for climbers
1. Recommendations for climbers
Presenter: Ivanova Iustina, PhD student at Free University of Bozen-Bolzano
Supervisor: Francesco Ricci
Co-supervisor: Marina Andric
2021
2. Content of the presentation
• Introduction
• Recommendations for climbers
• Activity recognition and recommendations
• Related work
• Design proposal for climbing recommendations
• Free software employed for the project
• Future plans
2/22
3. Introduction to sport climbing
• Sport climbing is becoming a popular sport
• Climbers are interested to climb in different places around the world
• There is a lack of an electronic support for climbers - electronic
climbing guidebooks do not offer any suggestions
3/22
https://www.climbingbusinessjournal.com/climbing-gyms-and-trends-2019/
4. Recommendations for climbers
• Create an innovative and useful decision support tools for climbers
and gym owners
• Automatically detecting climbing activities (who climbed what) is
important
• Understanding what climbers are doing at a certain point is an initial
step for modeling behaviour of climbers and for supporting them in
their decision
• This could be done via implicit or explicit feedback
4/22
6. Activity recognition in sport climbing
1) Pansiot et al. “ClimbBSN: Climber Performance Monitoring with BSN”, 2008
2) Kosmalla et al. “ClimbSense – Automatic Climbing Route Recognition using Wrist-worn
Inertia Measurement Units”, 2015
3) Ladha et al. “ClimbAX: Skill Assessment for Climbing Enthusiasts”, 2013
4) Seifert et al. “Full-body movement pattern recognition in climbing”, 2015
ClimBSN: 1 sensor[1]
Climbsense: 2 sensors[2]
5 sensors[4]
ClimbAX: 2 sensors[3]
6/22
7. 7/22
Related work: recommendations for training
in sport climbing
Some projects proposed to employ video cameras for training
betaCube
1) F. Wiehr et al. “betaCube – Enhancing Training for Climbing by a Self-Calibrating Camera-
Projection Unit”, 2016
2) K. Shiro et al. “InterPoser: Visualizing Interpolated Movements for Bouldering Training”, 2019
InterPoser
8. 8/22
Existing recommender systems in other
sports
RUNNING: MARATHON
PREPARATION[1]
FITNESS
RECOMMENDATIONS[2]
HIKING[3] ICE-SKATING[4]
1) Berndsen et al. “Running with recommendation”, 2017
2) Piloni et al. “Recommender System Lets Coaches Identify and Help Athletes Who Begin Losing
Motivation”, 2018
3) Calbimonte et al. “A Platform for Difficulty Assessment and Recommendation of Hiking Trails”, 2021
4) Smyth et al. “Predicting the Personal-Best Times of Speed Skaters Using Case-Based Reasoning”,
2020
10. Can we do something similar for climbers?
• Climbers are interested to climb routes which are:
• Challenging, but within their capabilities;
• Enjoyable;
• Safe;
• Within specified time.
• One important aspect is the difficulty level of climbing routes
• The main goal is to understand climbers’ needs and to provide
recommendations according to their needs
10/22
15. Recommendations for the climbing routes
within the crag
Climbing routes
recommendations
for training
Climbing routes
recommendations
for motivation
Climbing routes
recommendations
for favourite moves
Climbing routes
recommendations
within the crag
15/22
16. Recommendations for training
• Climber 1 climbs styles
‘endurance’ and ‘cruxy’, but
usually avoids ‘technical’ and
‘athletic’ styles
• Climber 2 climbs all the styles
equally
• Recommendation for climber 1
should be routes with ‘technical’
and ‘athletic’ styles
16/22
17. Recommendations for motivation: grade
prediction model
This Photo by Unknown Author is licensed under CC BY-NC
Climber
Route setter
Climbing routes
Route №5 is level 6c
Route №5 is level 7a
Disagreement between route setters and climbers
17/22
19. 19/22
Two methods to predict perceived grade
model
We consider problem as a rating prediction task, where
grade difference between climber and route setter’s grade is
a special type of rating.
We chose the models, successfully applied for rating
prediction before.
•Linear regression
•Matrix factorization
21. Future activities
• To introduce implicit feedback (data coming from sensors) to improve
recommendations for climbing routes
• To evaluate climbers’ opinions about the recommender system
developed for the small subset of climbing routes (located in Arco,
Italy)
• To improve recommendations taking into consideration the real
feedback of the climbers
• To improve recommendations for training purpose
21/22