4th NURSE CARE ACTIVITY RECOGNITION CHALLENGE DATASETS

Submission Dates:
06/01/2022 to 10/29/2022
Citation Author(s):
Sozo
Inoue
Kyushu Institute of Technology
Defry
Hamdhana
Kyushu Institute of Technology
Christina
Garcia
Kyushu Institute of Technology
Haru
Kaneko
Kyushu Institute of Technology
Nazmun
Nahid
Kyushu Institute of Technology
Tahera
Hossain
Kyushu Institute of Technology
Sayeda Shamma
Alia
Kyushu Institute of Technology
Paula
Lago
Kyushu Institute of Technology
Submitted by:
Sozolab Kyutech
Last updated:
Thu, 06/30/2022 - 01:03
DOI:
10.21227/vchd-s336
Data Format:
License:
Creative Commons Attribution

Abstract 

The data we are providing this time is a part of the dataset which was used in our previous work, titled “Integrating Activity Recognition and Nursing Care Records: The System, Deployment, and a Verification Study”. The authors of this work proposed a theory that extending of start and end times of the activities can increase the prediction rate. The reason behind the theory is that many of the nurses provided the labels before or after completing an activity. In the paper, they verified and proved this theory. A part of this data set has also been used in last year's nurse care challenge(https://abc-research.github.io/nurse2021/). The summary paper is titled” Summary of the Third Nurse Care Activity Recognition Challenge - Can We Do from the Field Data? (http://dx.doi.org/10.1145/3460418.3479391)”

 

Instructions: 

The training and testing dataset contains accelerometer data and care record data of 5 users ( 8, 13, 14, 15, 25), which were collected on May and June, 2018. Training and testing data were separated in 70~30 ratio based on each user data.

The provided each user folder includes the accelerometer data files for whole time, training care record file and testing care record file. In training care record we have activity information but in testing care record we remove it. Your work is to predict these activities and generate files.

For further information, please visit our website: https://abc-research.github.io/challenge2022/

Comments

hj

There is a query about test data and submission file formatting. In the submission file, you have provided the year-month-date-hour as well as activity labels. There was a total of 28 activities presented in the dataset. But the submission file format you have provided in the dataset contains different no. of activities (i.e., 21, 16) for the different user_id. Please clarify that we have to submit the exact file format you provided or to generate a new file consisting of a total of 28 activities for each user. Because, some segments predict the activity label that are not presented in the submission file activities columns.