Challenge 2024
The Sussex-Huawei Locomotion Dataset is going to be employed in the sixth and last edition of the SHL Challenge. The results of the challenge will be presented at the HASCA Workshop at Ubicomp/ISWC 2024.
The goal of this year edition is to recognize 8 modes of locomotion and transportation (activities) in a user-independent and opportunistic manner based on motion sensor data. This is different from the 2018-2020 Challenges that aimed at transportation mode recognition from motion sensors, and the 2021 Challenge that aimed at recognition from GPS and radio sensors.
Challenge 2023
The Sussex-Huawei Locomotion Dataset is going to be employed in the fifth edition of the SHL Challenge. The results of the challenge will be presented at the HASCA Workshop at Ubicomp/ISWC 2023.
The goal of this year edition is to recognize 8 modes of locomotion and transportation (activities) in a user-independent manner based on motion and GPS sensor data. This is different from the 2018-2020 Challenges that aimed at transportation mode recognition from motion sensors, and the 2021 Challenge that aimed at recognition from GPS and radio sensors.
Challenge 2021
The Sussex-Huawei Locomotion Dataset is going to be employed in the fourth edition of the SHL Challenge. The results of the challenge will be presented at the HASCA Workshop at Ubicomp/ISWC 2021.
The goal of this year edition is to recognize 8 modes of locomotion and transportation (activities) in a user-independent manner based on radio-data, including GPS reception, GPS location, Wifi reception and GSM cell tower scans. This is different from the previous three years that aimed at transportation mode recognition from the motion sensors.
Here the results (F1 scores) of the top 3 teams:
- DD (75.40%)
- Guerrilla (74.26%)
- We_can_fly (63.28%)
Challenge 2020
The Sussex-Huawei Locomotion Dataset is going to be employed in the third edition of the SHL Challenge. The results of the challenge will be presented at the HASCA Workshop at Ubicomp 2020.
The goal of this machine learning/data science challenge is to recognize 8 modes of locomotion and transportation (activities) from the inertial sensor data of a smartphone in a user independent manner. More precisely, the goal is to recognize the user activity from data coming from two users (User 2 and User 3), but training the model using all the data release so far, which are produced by User 1.
Here the results (F1 scores) of the top 3 teams:
- We_Can_Fly: 88.74%
- IndRNN (78.98%)
- ThirdTimesACharm (77.89%)
Challenge 2019
The Sussex-Huawei Locomotion Dataset was employed in the second edition of the SHL Challenge. The results of the challenge were presented at the HASCA Workshop at Ubicomp 2019.
The goal of this machine learning/data science challenge is to recognize 8 modes of locomotion and transportation (activities) from the inertial sensor data of a smartphone in a mobile-phone placement independent manner. More precisely, the goal is to recognize the user activity from data from the Hand phone, but training the model using data from smartphones on other different positions. A little validation data from the Hand phones is also provided. The dataset used for this challenge comprises 59 days of training data, 3 days of validation data, and 20 days of test data.
Here the results (F1 scores) of the top 3 teams:
- JSI-First: 78.42%
- Yonsei-MCML: 75.88%
- We_Can_Fly: 70.30%
Challenge 2018
The Sussex-Huawei Locomotion Dataset was used in the first edition of an activity recognition challenge of the same name. The results were presented at the HASCA Workshop at Ubicomp 2018.
The goal of this machine learning/data science challenge was to recognize 8 modes of locomotion and transportation (activities) from the inertial sensor data of a smartphone. The dataset used for this challenge comprises 271 hours of training data and 95 hours of test data, from a single phone on a single user.
Here the results (F1 scores) of the top 5 teams:
- JSI-Deep: 93.86%
- JSI-Classic: 92.41%
- Tesaguri: 88.83%
- S304: 87.46%
- Confusion Matrix: 87.45%