Smartwatch based activity recognition using active learning. Shahmohammadi, F., Hosseini, A., King, C. E., & Sarrafzadeh, M. In 2017 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies,, Philadelphia, PA, 2017. abstract bibtex —Human activity monitoring has become widely popular in recent years, and has been utilized in a vast number of fields and applications. Most of the activity recognition algorithms proposed have emphasized the use of inertial sensors in smartphone devices or other bodily-worn sensors. However, wearable inertial sensors are not interactive, and smartphones are not easily worn. Thus, with the advancement of smartwatches, unique opportunities exist to provide user interaction and highly accurate personalized activity recognition. Through the use of Active Learning, an interactive machine learning technique, specific behaviors can be learned by querying for unknown actions. This paper describes a smartwatch-based active learning method for activity recognition to identify 5 commonly performed daily activities. The results of this study revealed that this system can obtain a 93.3% accuracy across 12 participants. From our results, we demonstrate that an interactive learning approach using active learning in smartwatches has significant advantages over smartphones and other devices for activity recognition tasks.
@inproceedings{shahmohammadi_smartwatch_2017,
address = {Philadelphia, PA},
title = {Smartwatch based activity recognition using active learning.},
abstract = {—Human activity monitoring has become widely
popular in recent years, and has been utilized in a vast number of
fields and applications. Most of the activity recognition algorithms
proposed have emphasized the use of inertial sensors in
smartphone devices or other bodily-worn sensors. However,
wearable inertial sensors are not interactive, and smartphones are
not easily worn. Thus, with the advancement of smartwatches,
unique opportunities exist to provide user interaction and highly
accurate personalized activity recognition. Through the use of
Active Learning, an interactive machine learning technique,
specific behaviors can be learned by querying for unknown
actions. This paper describes a smartwatch-based active learning
method for activity recognition to identify 5 commonly performed
daily activities. The results of this study revealed that this system
can obtain a 93.3\% accuracy across 12 participants. From our
results, we demonstrate that an interactive learning approach
using active learning in smartwatches has significant advantages
over smartphones and other devices for activity recognition tasks.},
booktitle = {2017 {IEEE}/{ACM} {Conference} on {Connected} {Health}: {Applications}, {Systems} and {Engineering} {Technologies},},
author = {Shahmohammadi, F. and Hosseini, A. and King, C. E. and Sarrafzadeh, M.},
year = {2017},
}
Downloads: 0
{"_id":"PFzyXcBJH2XySYWrX","bibbaseid":"shahmohammadi-hosseini-king-sarrafzadeh-smartwatchbasedactivityrecognitionusingactivelearning-2017","author_short":["Shahmohammadi, F.","Hosseini, A.","King, C. E.","Sarrafzadeh, M."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"Philadelphia, PA","title":"Smartwatch based activity recognition using active learning.","abstract":"—Human activity monitoring has become widely popular in recent years, and has been utilized in a vast number of fields and applications. Most of the activity recognition algorithms proposed have emphasized the use of inertial sensors in smartphone devices or other bodily-worn sensors. However, wearable inertial sensors are not interactive, and smartphones are not easily worn. Thus, with the advancement of smartwatches, unique opportunities exist to provide user interaction and highly accurate personalized activity recognition. Through the use of Active Learning, an interactive machine learning technique, specific behaviors can be learned by querying for unknown actions. This paper describes a smartwatch-based active learning method for activity recognition to identify 5 commonly performed daily activities. The results of this study revealed that this system can obtain a 93.3% accuracy across 12 participants. From our results, we demonstrate that an interactive learning approach using active learning in smartwatches has significant advantages over smartphones and other devices for activity recognition tasks.","booktitle":"2017 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies,","author":[{"propositions":[],"lastnames":["Shahmohammadi"],"firstnames":["F."],"suffixes":[]},{"propositions":[],"lastnames":["Hosseini"],"firstnames":["A."],"suffixes":[]},{"propositions":[],"lastnames":["King"],"firstnames":["C.","E."],"suffixes":[]},{"propositions":[],"lastnames":["Sarrafzadeh"],"firstnames":["M."],"suffixes":[]}],"year":"2017","bibtex":"@inproceedings{shahmohammadi_smartwatch_2017,\n\taddress = {Philadelphia, PA},\n\ttitle = {Smartwatch based activity recognition using active learning.},\n\tabstract = {—Human activity monitoring has become widely\npopular in recent years, and has been utilized in a vast number of\nfields and applications. Most of the activity recognition algorithms\nproposed have emphasized the use of inertial sensors in\nsmartphone devices or other bodily-worn sensors. However,\nwearable inertial sensors are not interactive, and smartphones are\nnot easily worn. Thus, with the advancement of smartwatches,\nunique opportunities exist to provide user interaction and highly\naccurate personalized activity recognition. Through the use of\nActive Learning, an interactive machine learning technique,\nspecific behaviors can be learned by querying for unknown\nactions. This paper describes a smartwatch-based active learning\nmethod for activity recognition to identify 5 commonly performed\ndaily activities. The results of this study revealed that this system\ncan obtain a 93.3\\% accuracy across 12 participants. From our\nresults, we demonstrate that an interactive learning approach\nusing active learning in smartwatches has significant advantages\nover smartphones and other devices for activity recognition tasks.},\n\tbooktitle = {2017 {IEEE}/{ACM} {Conference} on {Connected} {Health}: {Applications}, {Systems} and {Engineering} {Technologies},},\n\tauthor = {Shahmohammadi, F. and Hosseini, A. and King, C. E. and Sarrafzadeh, M.},\n\tyear = {2017},\n}\n\n","author_short":["Shahmohammadi, F.","Hosseini, A.","King, C. E.","Sarrafzadeh, M."],"key":"shahmohammadi_smartwatch_2017","id":"shahmohammadi_smartwatch_2017","bibbaseid":"shahmohammadi-hosseini-king-sarrafzadeh-smartwatchbasedactivityrecognitionusingactivelearning-2017","role":"author","urls":{},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://api.zotero.org/users/3649949/collections/52M3HD2M/items?key=kvw05jEWpV9zO4gNkD1KQFRV&format=bibtex&limit=100&jsonp=1&filter=type:misc","dataSources":["qkN2F4hKQojRGQeTy","cR2bQCnuvgoCQwTEh","Zv8utRXNjhXZcJdZX","zguJ5LkMpKLhRDgWX","x94sDkjv6sHRisXm3"],"keywords":[],"search_terms":["smartwatch","based","activity","recognition","using","active","learning","shahmohammadi","hosseini","king","sarrafzadeh"],"title":"Smartwatch based activity recognition using active learning.","year":2017}