An optimized cosine similarity-based attention gate for temporal sequence patterns recognition
The attention gate is often employed in various applications, such as recommender systems, to determine the correct context of the input data. Adopting the attention gate has been proven to improve prediction accuracy successfully. However, for a temporal sequence problem such as sensor-based Sign L...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/107881/1/AmmarAbdullah2023_AnOptimizedCosineSimilarityBasedAttention.pdf |
| _version_ | 1817854197373075456 |
|---|---|
| author | Abdullah, Ammar Abdul Kadir, Nurul Ashikin Che Harun, Fauzan Khairi |
| author_facet | Abdullah, Ammar Abdul Kadir, Nurul Ashikin Che Harun, Fauzan Khairi |
| author_sort | Abdullah, Ammar |
| collection | ePrints |
| description | The attention gate is often employed in various applications, such as recommender systems, to determine the correct context of the input data. Adopting the attention gate has been proven to improve prediction accuracy successfully. However, for a temporal sequence problem such as sensor-based Sign Language Recognition (SLR), it is still challenging to integrate the attention gate into the solution since the processing input is only in the form of a sine waveform. In our work, we propose an optimized cosine similarity-based attention gate to manipulate the sine wave signal while improving the recognition of temporal sequence data. We also evaluate and compare the various distance calculations and determine the best one for the SLR application. Adopting a distance-based attention gate has successfully achieved the recognition accuracy of 99% while reducing the error rate to 5%. |
| first_indexed | 2024-12-08T06:54:18Z |
| format | Conference or Workshop Item |
| id | utm.eprints-107881 |
| institution | Universiti Teknologi Malaysia - ePrints |
| language | English |
| last_indexed | 2024-12-08T06:54:18Z |
| publishDate | 2023 |
| record_format | dspace |
| spelling | utm.eprints-1078812024-10-08T06:51:01Z http://eprints.utm.my/107881/ An optimized cosine similarity-based attention gate for temporal sequence patterns recognition Abdullah, Ammar Abdul Kadir, Nurul Ashikin Che Harun, Fauzan Khairi TK Electrical engineering. Electronics Nuclear engineering The attention gate is often employed in various applications, such as recommender systems, to determine the correct context of the input data. Adopting the attention gate has been proven to improve prediction accuracy successfully. However, for a temporal sequence problem such as sensor-based Sign Language Recognition (SLR), it is still challenging to integrate the attention gate into the solution since the processing input is only in the form of a sine waveform. In our work, we propose an optimized cosine similarity-based attention gate to manipulate the sine wave signal while improving the recognition of temporal sequence data. We also evaluate and compare the various distance calculations and determine the best one for the SLR application. Adopting a distance-based attention gate has successfully achieved the recognition accuracy of 99% while reducing the error rate to 5%. 2023 Conference or Workshop Item PeerReviewed application/pdf en http://eprints.utm.my/107881/1/AmmarAbdullah2023_AnOptimizedCosineSimilarityBasedAttention.pdf Abdullah, Ammar and Abdul Kadir, Nurul Ashikin and Che Harun, Fauzan Khairi (2023) An optimized cosine similarity-based attention gate for temporal sequence patterns recognition. In: 1st International Conference on Electronic and Computer Engineering, ECE 2023, 4 July 2023 - 5 July 2023, Virtual, UTM Johor Bahru, Johor, Malaysia. http://dx.doi.org/10.1088/1742-6596/2622/1/012012 |
| spellingShingle | TK Electrical engineering. Electronics Nuclear engineering Abdullah, Ammar Abdul Kadir, Nurul Ashikin Che Harun, Fauzan Khairi An optimized cosine similarity-based attention gate for temporal sequence patterns recognition |
| title | An optimized cosine similarity-based attention gate for temporal sequence patterns recognition |
| title_full | An optimized cosine similarity-based attention gate for temporal sequence patterns recognition |
| title_fullStr | An optimized cosine similarity-based attention gate for temporal sequence patterns recognition |
| title_full_unstemmed | An optimized cosine similarity-based attention gate for temporal sequence patterns recognition |
| title_short | An optimized cosine similarity-based attention gate for temporal sequence patterns recognition |
| title_sort | optimized cosine similarity based attention gate for temporal sequence patterns recognition |
| topic | TK Electrical engineering. Electronics Nuclear engineering |
| url | http://eprints.utm.my/107881/1/AmmarAbdullah2023_AnOptimizedCosineSimilarityBasedAttention.pdf |
| work_keys_str_mv | AT abdullahammar anoptimizedcosinesimilaritybasedattentiongatefortemporalsequencepatternsrecognition AT abdulkadirnurulashikin anoptimizedcosinesimilaritybasedattentiongatefortemporalsequencepatternsrecognition AT cheharunfauzankhairi anoptimizedcosinesimilaritybasedattentiongatefortemporalsequencepatternsrecognition AT abdullahammar optimizedcosinesimilaritybasedattentiongatefortemporalsequencepatternsrecognition AT abdulkadirnurulashikin optimizedcosinesimilaritybasedattentiongatefortemporalsequencepatternsrecognition AT cheharunfauzankhairi optimizedcosinesimilaritybasedattentiongatefortemporalsequencepatternsrecognition |