Article Details

Abstract

In the era of information explosion, the problem of information overload in online news is becoming increasingly serious, and traditional classification methods often fail to accurately identify key information in multimodal scenarios. To address these issues, a multimodal-based keyword extraction method is designed and proposed. By means of semantic understanding of long texts and extraction of key information from news images, this method fuses and screens the key information in multimodal news content, and improves the accuracy of news keyword extraction through multimodal fusion. The results of comparative experiments with multiple baseline models demonstrate that the designed model can achieve better performance indicators in multimodal keyword extraction scenarios, and the keyword extraction results are more accurate.

Keywords

keyword extraction; multimodal fusion; pre-trained model; object recognition; semantic fusion embedding

References

1. [1] Tang Yu Research and Improvement of Information Retrieval Algorithms [D]. Nanjing: Nanjing University of Posts and Telecommunications, 2022
2. [2] Feng Yue Research on Forestry News Recommendation System Based on Personalized Recommendation [D]. Beijing: Beijing Forestry University, 2022
3. [3] Lu Huihong, Ma Ping, Wang Xiao Virtual anchor comment sentiment classification algorithm based on BERT-CLS-ATT model [J]. Journal of People's Public Security University of China: Natural Science Edition, 2024, 30 (1): 74-80
4. [4] Su Jingqiong, Su Yanqiong Research on Keyword Extraction Algorithm Based on LDA and TF-IDF [J]. Changjiang Information and Communication, 2024, 37 (1): 78-80
5. [5] Lan Xiaofang, Liu Zhuo, Xu Zhihao, etc Chinese Text Keyword Extraction Method Based on TF-IDF and TextRank: A Case Study of Sports News [J]. Software Engineering, 2023, 26 (8): 6-10
6. [6] PAN R. Automatic Keyword Extraction Algorithm for Chinese Text based on Word Clustering [J/OL].Transactions on Asian and Low-Resource Language Information Processing，2023：1-17（2023-04-08）.https://dl.acm.org/doi/10.1145/3592793.
7. [7] SARWAR T B，NOOR N K M，MIAH M. Evaluating Keyphrase Extraction Algorithms for Finding Similar News Articles Using Lexical Similarity Calculation and Semantic Relatedness Measurement by Word Embedding [J/OL].PeerJ Computer Science，2022（8）：e1024（2022-07-07）.https://peerj.com/articles/cs-1024/.
8. [8] ESPADA J P，MARTINEZ J S，RICO I C，et al. Extracting Keywords of Educational texts Using a Novel Mechanism based on Linguistic Approaches and Evolutive Graphs [J/OL].Expert Systems with Applications，2023：118842（2022-10-01）.https://doi.org/10.1016/j.eswa.2022.118842.
9. [9] Wang Yuye, Wang Di Research on Patent Keyword Extraction Algorithm Based on Graph Neural Network [J]. Intelligence Theory and Practice, 2023, 46 (5): 2022-208
10. [10] ABIBULLAYEVA A，HÜMA K，CETIN A. An Ensemble Keyword Extraction Model for News Texts with Statistical and Graphical Features [J].International Journal of Software Engineering and Knowledge Engineering，2024，34（7）：1047-1061.
11. [11] MENG R，ZHAO S，HAN S，et al. Deep Keyphrase Generation [C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).Vancouver，ACL，2017：582-592.
12. [12] YUAN X，WANG T，MENG R，et al. One Size Does Not Fit All: Generating and Evaluating Variable Number of Keyphrases [C/OL]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.ACL，2020：7961-7975[2024-09-10].https://aclanthology.org/2020.acl-main.710/.
13. [13] GUO L，SUN H，QI Q，et al. Keyword Extraction Algorithm Based on Pre-training and Multi-task Training [C]//Proceedings of Sixth International Congress on Information and Communication Technology.London：Springer，2022：723-734.
14. [14] MA J，CHENG J，ZHANG Y. A Novel Keyword Generation Model Based on Topic-Aware and Title-Guide [J/OL].Computational Intelligence and Neuroscience，2022：1-9（2022-01-01）.https://doi.org/10.1155/2022/1787369
15. [15] CHEN W，CHAN H P，LI P，et al. Exclusive Hierarchical Decoding for Deep Keyphrase Generation [C/OL]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020：1095-1105[2024-09-10].https://aclanthology.org/2020.acl-main.103/.

About the authors

Liu Yu (1978-), male, Han ethnicity, from Wuhu, Anhui, is a senior engineer with a master's degree. His research interests include big data and news recommendation.