Intent Detection and Slot Filling with Capsule Net Architectures for a Romanian Home Assistant
As virtual home assistants are becoming more popular, there is an emerging need for supporting languages other than English. While more wide-spread or popular languages such as Spanish, French or Hindi are already integrated into existing home assistants like Google Home or Alexa, integration of oth...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/4/1230 |
| _version_ | 1797412073165029376 |
|---|---|
| author | Anda Stoica Tibor Kadar Camelia Lemnaru Rodica Potolea Mihaela Dînşoreanu |
| author_facet | Anda Stoica Tibor Kadar Camelia Lemnaru Rodica Potolea Mihaela Dînşoreanu |
| author_sort | Anda Stoica |
| collection | DOAJ |
| description | As virtual home assistants are becoming more popular, there is an emerging need for supporting languages other than English. While more wide-spread or popular languages such as Spanish, French or Hindi are already integrated into existing home assistants like Google Home or Alexa, integration of other less-known languages such as Romanian is still missing. This paper explores the problem of Natural Language Understanding (NLU) applied to a Romanian home assistant. We propose a customized capsule neural network architecture that performs intent detection and slot filling in a joint manner and we evaluate how well it handles utterances containing various levels of complexity. The capsule network model shows a significant improvement in intent detection when compared to models built using the well-known Rasa NLU tool. Through error analysis, we observe clear error patterns that occur systematically. Variability in language when expressing one intent proves to be the biggest challenge encountered by the model. |
| first_indexed | 2024-03-09T04:55:44Z |
| format | Article |
| id | doaj.art-07c1afb6578840cb92d2aa6e2029f1fe |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T04:55:44Z |
| publishDate | 2021-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-07c1afb6578840cb92d2aa6e2029f1fe2023-12-03T13:05:08ZengMDPI AGSensors1424-82202021-02-01214123010.3390/s21041230Intent Detection and Slot Filling with Capsule Net Architectures for a Romanian Home AssistantAnda Stoica0Tibor Kadar1Camelia Lemnaru2Rodica Potolea3Mihaela Dînşoreanu4Department of Computer Science, Technical University of Cluj-Napoca, 26-28 G. Baritiu, 400027 Cluj-Napoca, RomaniaDepartment of Computer Science, Technical University of Cluj-Napoca, 26-28 G. Baritiu, 400027 Cluj-Napoca, RomaniaDepartment of Computer Science, Technical University of Cluj-Napoca, 26-28 G. Baritiu, 400027 Cluj-Napoca, RomaniaDepartment of Computer Science, Technical University of Cluj-Napoca, 26-28 G. Baritiu, 400027 Cluj-Napoca, RomaniaDepartment of Computer Science, Technical University of Cluj-Napoca, 26-28 G. Baritiu, 400027 Cluj-Napoca, RomaniaAs virtual home assistants are becoming more popular, there is an emerging need for supporting languages other than English. While more wide-spread or popular languages such as Spanish, French or Hindi are already integrated into existing home assistants like Google Home or Alexa, integration of other less-known languages such as Romanian is still missing. This paper explores the problem of Natural Language Understanding (NLU) applied to a Romanian home assistant. We propose a customized capsule neural network architecture that performs intent detection and slot filling in a joint manner and we evaluate how well it handles utterances containing various levels of complexity. The capsule network model shows a significant improvement in intent detection when compared to models built using the well-known Rasa NLU tool. Through error analysis, we observe clear error patterns that occur systematically. Variability in language when expressing one intent proves to be the biggest challenge encountered by the model.https://www.mdpi.com/1424-8220/21/4/1230NLUintent detectionslot fillingcapsule neural networksRomanian home assistant |
| spellingShingle | Anda Stoica Tibor Kadar Camelia Lemnaru Rodica Potolea Mihaela Dînşoreanu Intent Detection and Slot Filling with Capsule Net Architectures for a Romanian Home Assistant Sensors NLU intent detection slot filling capsule neural networks Romanian home assistant |
| title | Intent Detection and Slot Filling with Capsule Net Architectures for a Romanian Home Assistant |
| title_full | Intent Detection and Slot Filling with Capsule Net Architectures for a Romanian Home Assistant |
| title_fullStr | Intent Detection and Slot Filling with Capsule Net Architectures for a Romanian Home Assistant |
| title_full_unstemmed | Intent Detection and Slot Filling with Capsule Net Architectures for a Romanian Home Assistant |
| title_short | Intent Detection and Slot Filling with Capsule Net Architectures for a Romanian Home Assistant |
| title_sort | intent detection and slot filling with capsule net architectures for a romanian home assistant |
| topic | NLU intent detection slot filling capsule neural networks Romanian home assistant |
| url | https://www.mdpi.com/1424-8220/21/4/1230 |
| work_keys_str_mv | AT andastoica intentdetectionandslotfillingwithcapsulenetarchitecturesforaromanianhomeassistant AT tiborkadar intentdetectionandslotfillingwithcapsulenetarchitecturesforaromanianhomeassistant AT camelialemnaru intentdetectionandslotfillingwithcapsulenetarchitecturesforaromanianhomeassistant AT rodicapotolea intentdetectionandslotfillingwithcapsulenetarchitecturesforaromanianhomeassistant AT mihaeladinsoreanu intentdetectionandslotfillingwithcapsulenetarchitecturesforaromanianhomeassistant |