Natural language understanding of map navigation queries in Roman Urdu by joint entity and intent determination
Navigation based task-oriented dialogue systems provide users with a natural way of communicating with maps and navigation software. Natural language understanding (NLU) is the first step for a task-oriented dialogue system. It extracts the important entities (slot tagging) from the user’s utterance...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2021-07-01
|
| Series: | PeerJ Computer Science |
| Subjects: | |
| Online Access: | https://peerj.com/articles/cs-615.pdf |
| _version_ | 1819148709244436480 |
|---|---|
| author | Javeria Hassan Muhammad Ali Tahir Adnan Ali |
| author_facet | Javeria Hassan Muhammad Ali Tahir Adnan Ali |
| author_sort | Javeria Hassan |
| collection | DOAJ |
| description | Navigation based task-oriented dialogue systems provide users with a natural way of communicating with maps and navigation software. Natural language understanding (NLU) is the first step for a task-oriented dialogue system. It extracts the important entities (slot tagging) from the user’s utterance and determines the user’s objective (intent determination). Word embeddings are the distributed representations of the input sentence, and encompass the sentence’s semantic and syntactic representations. We created the word embeddings using different methods like FastText, ELMO, BERT and XLNET; and studied their effect on the natural language understanding output. Experiments are performed on the Roman Urdu navigation utterances dataset. The results show that for the intent determination task XLNET based word embeddings outperform other methods; while for the task of slot tagging FastText and XLNET based word embeddings have much better accuracy in comparison to other approaches. |
| first_indexed | 2024-12-22T13:50:01Z |
| format | Article |
| id | doaj.art-c1acec6387ba4edc88773d6c9c5a2579 |
| institution | Directory Open Access Journal |
| issn | 2376-5992 |
| language | English |
| last_indexed | 2024-12-22T13:50:01Z |
| publishDate | 2021-07-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ Computer Science |
| spelling | doaj.art-c1acec6387ba4edc88773d6c9c5a25792022-12-21T18:23:41ZengPeerJ Inc.PeerJ Computer Science2376-59922021-07-017e61510.7717/peerj-cs.615Natural language understanding of map navigation queries in Roman Urdu by joint entity and intent determinationJaveria Hassan0Muhammad Ali Tahir1Adnan Ali2National University of Sciences and Technology (NUST), Islamabad, PakistanNational University of Sciences and Technology (NUST), Islamabad, PakistanUniversity of Science and Technology of China, Hefei, Anhui, ChinaNavigation based task-oriented dialogue systems provide users with a natural way of communicating with maps and navigation software. Natural language understanding (NLU) is the first step for a task-oriented dialogue system. It extracts the important entities (slot tagging) from the user’s utterance and determines the user’s objective (intent determination). Word embeddings are the distributed representations of the input sentence, and encompass the sentence’s semantic and syntactic representations. We created the word embeddings using different methods like FastText, ELMO, BERT and XLNET; and studied their effect on the natural language understanding output. Experiments are performed on the Roman Urdu navigation utterances dataset. The results show that for the intent determination task XLNET based word embeddings outperform other methods; while for the task of slot tagging FastText and XLNET based word embeddings have much better accuracy in comparison to other approaches.https://peerj.com/articles/cs-615.pdfNatural language understandingRoman UrduNavigation queryWord embeddings |
| spellingShingle | Javeria Hassan Muhammad Ali Tahir Adnan Ali Natural language understanding of map navigation queries in Roman Urdu by joint entity and intent determination PeerJ Computer Science Natural language understanding Roman Urdu Navigation query Word embeddings |
| title | Natural language understanding of map navigation queries in Roman Urdu by joint entity and intent determination |
| title_full | Natural language understanding of map navigation queries in Roman Urdu by joint entity and intent determination |
| title_fullStr | Natural language understanding of map navigation queries in Roman Urdu by joint entity and intent determination |
| title_full_unstemmed | Natural language understanding of map navigation queries in Roman Urdu by joint entity and intent determination |
| title_short | Natural language understanding of map navigation queries in Roman Urdu by joint entity and intent determination |
| title_sort | natural language understanding of map navigation queries in roman urdu by joint entity and intent determination |
| topic | Natural language understanding Roman Urdu Navigation query Word embeddings |
| url | https://peerj.com/articles/cs-615.pdf |
| work_keys_str_mv | AT javeriahassan naturallanguageunderstandingofmapnavigationqueriesinromanurdubyjointentityandintentdetermination AT muhammadalitahir naturallanguageunderstandingofmapnavigationqueriesinromanurdubyjointentityandintentdetermination AT adnanali naturallanguageunderstandingofmapnavigationqueriesinromanurdubyjointentityandintentdetermination |