Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes
Long Short Term Memory (LSTM)-based structures have demonstrated their efficiency for daily living recognition activities in smart homes by capturing the order of sensor activations and their temporal dependencies. Nevertheless, they still fail in dealing with the semantics and the context of the se...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-10-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/10/20/2498 |
| _version_ | 1797514796564742144 |
|---|---|
| author | Damien Bouchabou Sao Mai Nguyen Christophe Lohr Benoit LeDuc Ioannis Kanellos |
| author_facet | Damien Bouchabou Sao Mai Nguyen Christophe Lohr Benoit LeDuc Ioannis Kanellos |
| author_sort | Damien Bouchabou |
| collection | DOAJ |
| description | Long Short Term Memory (LSTM)-based structures have demonstrated their efficiency for daily living recognition activities in smart homes by capturing the order of sensor activations and their temporal dependencies. Nevertheless, they still fail in dealing with the semantics and the context of the sensors. More than isolated id and their ordered activation values, sensors also carry meaning. Indeed, their nature and type of activation can translate various activities. Their logs are correlated with each other, creating a global context. We propose to use and compare two Natural Language Processing embedding methods to enhance LSTM-based structures in activity-sequences classification tasks: Word2Vec, a static semantic embedding, and ELMo, a contextualized embedding. Results, on real smart homes datasets, indicate that this approach provides useful information, such as a sensor organization map, and makes less confusion between daily activity classes. It helps to better perform on datasets with competing activities of other residents or pets. Our tests show also that the embeddings can be pretrained on different datasets than the target one, enabling transfer learning. We thus demonstrate that taking into account the context of the sensors and their semantics increases the classification performances and enables transfer learning. |
| first_indexed | 2024-03-10T06:36:39Z |
| format | Article |
| id | doaj.art-79c2caf1a7744932852c470eac66255b |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T06:36:39Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-79c2caf1a7744932852c470eac66255b2023-11-22T18:02:11ZengMDPI AGElectronics2079-92922021-10-011020249810.3390/electronics10202498Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart HomesDamien Bouchabou0Sao Mai Nguyen1Christophe Lohr2Benoit LeDuc3Ioannis Kanellos4IMT Atlantique Engineer School, 29238 Brest, FranceIMT Atlantique Engineer School, 29238 Brest, FranceIMT Atlantique Engineer School, 29238 Brest, FranceDelta Dore Company, 35270 Bonnemain, FranceIMT Atlantique Engineer School, 29238 Brest, FranceLong Short Term Memory (LSTM)-based structures have demonstrated their efficiency for daily living recognition activities in smart homes by capturing the order of sensor activations and their temporal dependencies. Nevertheless, they still fail in dealing with the semantics and the context of the sensors. More than isolated id and their ordered activation values, sensors also carry meaning. Indeed, their nature and type of activation can translate various activities. Their logs are correlated with each other, creating a global context. We propose to use and compare two Natural Language Processing embedding methods to enhance LSTM-based structures in activity-sequences classification tasks: Word2Vec, a static semantic embedding, and ELMo, a contextualized embedding. Results, on real smart homes datasets, indicate that this approach provides useful information, such as a sensor organization map, and makes less confusion between daily activity classes. It helps to better perform on datasets with competing activities of other residents or pets. Our tests show also that the embeddings can be pretrained on different datasets than the target one, enabling transfer learning. We thus demonstrate that taking into account the context of the sensors and their semantics increases the classification performances and enables transfer learning.https://www.mdpi.com/2079-9292/10/20/2498sensors embeddinghuman activity recognitiondeep learningsmart homeambient assisting livinglanguage model |
| spellingShingle | Damien Bouchabou Sao Mai Nguyen Christophe Lohr Benoit LeDuc Ioannis Kanellos Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes Electronics sensors embedding human activity recognition deep learning smart home ambient assisting living language model |
| title | Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes |
| title_full | Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes |
| title_fullStr | Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes |
| title_full_unstemmed | Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes |
| title_short | Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes |
| title_sort | using language model to bootstrap human activity recognition ambient sensors based in smart homes |
| topic | sensors embedding human activity recognition deep learning smart home ambient assisting living language model |
| url | https://www.mdpi.com/2079-9292/10/20/2498 |
| work_keys_str_mv | AT damienbouchabou usinglanguagemodeltobootstraphumanactivityrecognitionambientsensorsbasedinsmarthomes AT saomainguyen usinglanguagemodeltobootstraphumanactivityrecognitionambientsensorsbasedinsmarthomes AT christophelohr usinglanguagemodeltobootstraphumanactivityrecognitionambientsensorsbasedinsmarthomes AT benoitleduc usinglanguagemodeltobootstraphumanactivityrecognitionambientsensorsbasedinsmarthomes AT ioanniskanellos usinglanguagemodeltobootstraphumanactivityrecognitionambientsensorsbasedinsmarthomes |