News Topic Detection Based on Capsule Semantic Graph
Most news topic detection methods use word-based methods, which easily ignore the relationship among words and have semantic sparsity, resulting in low topic detection accuracy. In addition, the current mainstream probability methods and graph analysis methods for topic detection have high time comp...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2022-06-01
|
| Series: | Big Data Mining and Analytics |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.26599/BDMA.2021.9020023 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832572923555086336 |
|---|---|
| author | Shuang Yang Yan Tang |
| author_facet | Shuang Yang Yan Tang |
| author_sort | Shuang Yang |
| collection | DOAJ |
| description | Most news topic detection methods use word-based methods, which easily ignore the relationship among words and have semantic sparsity, resulting in low topic detection accuracy. In addition, the current mainstream probability methods and graph analysis methods for topic detection have high time complexity. For these reasons, we present a news topic detection model on the basis of capsule semantic graph (CSG). The keywords that appear in each text at the same time are modeled as a keyword graph, which is divided into multiple subgraphs through community detection. Each subgraph contains a group of closely related keywords. The graph is used as the vertex of CSG. The semantic relationship among the vertices is obtained by calculating the similarity of the average word vector of each vertex. At the same time, the news text is clustered using the incremental clustering method, where each text uses CSG; that is, the similarity among texts is calculated by the graph kernel. The relationship between vertices and edges is also considered when calculating the similarity. Experimental results on three standard datasets show that CSG can obtain higher precision, recall, and F1 values than several latest methods. Experimental results on large-scale news datasets reveal that the time complexity of CSG is lower than that of probabilistic methods and other graph analysis methods. |
| format | Article |
| id | doaj-art-0f29e6bae1bb40b2bd24506ec9964adf |
| institution | Kabale University |
| issn | 2096-0654 |
| language | English |
| publishDate | 2022-06-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Big Data Mining and Analytics |
| spelling | doaj-art-0f29e6bae1bb40b2bd24506ec9964adf2025-02-02T06:14:03ZengTsinghua University PressBig Data Mining and Analytics2096-06542022-06-01529810910.26599/BDMA.2021.9020023News Topic Detection Based on Capsule Semantic GraphShuang Yang0Yan Tang1College of Computer and Information Science, Southwest University, Chongqing 400000, ChinaCollege of Computer and Information Science, Southwest University, Chongqing 400000, ChinaMost news topic detection methods use word-based methods, which easily ignore the relationship among words and have semantic sparsity, resulting in low topic detection accuracy. In addition, the current mainstream probability methods and graph analysis methods for topic detection have high time complexity. For these reasons, we present a news topic detection model on the basis of capsule semantic graph (CSG). The keywords that appear in each text at the same time are modeled as a keyword graph, which is divided into multiple subgraphs through community detection. Each subgraph contains a group of closely related keywords. The graph is used as the vertex of CSG. The semantic relationship among the vertices is obtained by calculating the similarity of the average word vector of each vertex. At the same time, the news text is clustered using the incremental clustering method, where each text uses CSG; that is, the similarity among texts is calculated by the graph kernel. The relationship between vertices and edges is also considered when calculating the similarity. Experimental results on three standard datasets show that CSG can obtain higher precision, recall, and F1 values than several latest methods. Experimental results on large-scale news datasets reveal that the time complexity of CSG is lower than that of probabilistic methods and other graph analysis methods.https://www.sciopen.com/article/10.26599/BDMA.2021.9020023news topic detectioncapsule semantic graphgraph kernel |
| spellingShingle | Shuang Yang Yan Tang News Topic Detection Based on Capsule Semantic Graph Big Data Mining and Analytics news topic detection capsule semantic graph graph kernel |
| title | News Topic Detection Based on Capsule Semantic Graph |
| title_full | News Topic Detection Based on Capsule Semantic Graph |
| title_fullStr | News Topic Detection Based on Capsule Semantic Graph |
| title_full_unstemmed | News Topic Detection Based on Capsule Semantic Graph |
| title_short | News Topic Detection Based on Capsule Semantic Graph |
| title_sort | news topic detection based on capsule semantic graph |
| topic | news topic detection capsule semantic graph graph kernel |
| url | https://www.sciopen.com/article/10.26599/BDMA.2021.9020023 |
| work_keys_str_mv | AT shuangyang newstopicdetectionbasedoncapsulesemanticgraph AT yantang newstopicdetectionbasedoncapsulesemanticgraph |