WEKO3
アイテム
A Quality Analysis of the BiLSTM Encoder Decoder Model with Modified J-Divergence for Sentences with Different Complexities
http://hdl.handle.net/10228/0002001337
http://hdl.handle.net/10228/000200133747adab88-5dc4-4a6b-b0ea-661bcab9417f
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
10448549.pdf (2.1 MB)
|
|
| アイテムタイプ | 共通アイテムタイプ(1) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 公開日 | 2025-02-17 | |||||||||||
| タイトル | ||||||||||||
| タイトル | A Quality Analysis of the BiLSTM Encoder Decoder Model with Modified J-Divergence for Sentences with Different Complexities | |||||||||||
| 言語 | en | |||||||||||
| その他のタイトル | ||||||||||||
| その他のタイトル | A QUALITY ANALYSIS OF THE BILSTM ENCODER DECODER MODEL WITH MODIFIED J-DIVERGENCE FOR SENTENCES WITH DIFFERENT COMPLEXITIES | |||||||||||
| 言語 | en | |||||||||||
| 著者 |
Shrivastava, Manu
× Shrivastava, Manu
× 我妻, 広明
WEKO
30799
|
|||||||||||
| 著作権関連情報 | ||||||||||||
| 権利情報 | ICIC International Copyright (c) 2024 | |||||||||||
| 抄録 | ||||||||||||
| 内容記述タイプ | Abstract | |||||||||||
| 内容記述 | The rapid increase in the field of artificial intelligence has made machines capable of making intelligent decisions. Machines can now recognize patterns and support human intelligent activities, and one of the main sources for machines to learn is natural languages. Natural languages are complex and sentences can have different structures, difficult for a machine to understand. A formal way to represent language is ontology, and ontology organizes information as a triple and set of rules. The task of extracting triples from various sources such as text documents, or web pages is termed as ontology population task that can be formulated either as a classification or a translation task. In both these tasks data from an unknown probability distribution is fed to a neural network that generates another probability distribution parametrized by network weight, and the objective here is to minimize the distance between two probability distributions by minimizing some loss function. In this research, we proposed a modified version of Jeffreys divergence as a learning method for machines to improve the ontology population task. We also analyze the performance of the neural network model based on different structures of sentences used for training and propose data selection method for improving model performance. | |||||||||||
| 言語 | en | |||||||||||
| 書誌情報 |
en : ICIC Express Letters: An international journal of research and surveys 巻 18, 号 7, p. 721-729, 発行日 2024-07 |
|||||||||||
| 出版社 | ||||||||||||
| 出版者 | ICIC International | |||||||||||
| 言語 | en | |||||||||||
| キーワード | ||||||||||||
| 主題Scheme | Other | |||||||||||
| 主題 | KL divergence | |||||||||||
| キーワード | ||||||||||||
| 主題Scheme | Other | |||||||||||
| 主題 | Jeffreys divergence | |||||||||||
| キーワード | ||||||||||||
| 主題Scheme | Other | |||||||||||
| 主題 | RDF | |||||||||||
| キーワード | ||||||||||||
| 主題Scheme | Other | |||||||||||
| 主題 | Information retrieval | |||||||||||
| キーワード | ||||||||||||
| 主題Scheme | Other | |||||||||||
| 主題 | Sentence structure | |||||||||||
| 言語 | ||||||||||||
| 言語 | eng | |||||||||||
| 資源タイプ | ||||||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||||||||
| 資源タイプ | journal article | |||||||||||
| 出版タイプ | ||||||||||||
| 出版タイプ | VoR | |||||||||||
| 出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||||||||
| DOI | ||||||||||||
| 識別子タイプ | DOI | |||||||||||
| 関連識別子 | https://doi.org/10.24507/icicel.18.07.721 | |||||||||||
| ISSN | ||||||||||||
| 収録物識別子タイプ | PISSN | |||||||||||
| 収録物識別子 | 1881-803X | |||||||||||
| 研究者情報 | ||||||||||||
| URL | https://hyokadb02.jimu.kyutech.ac.jp/html/358_ja.html | |||||||||||
| 論文ID（連携） | ||||||||||||
| 値 | 10448549 | |||||||||||
| 連携ID | ||||||||||||
| 値 | 13018 | |||||||||||
