Search

Your search keyword '"Deng, Shumin"' showing total 274 results
Start Over Author "Deng, Shumin"

Refine your results

274 results on '"Deng, Shumin"'

102. Knowledge Extraction in Low-Resource Scenarios: Survey and Perspective

Author

Deng, Shumin, Zhang, Ningyu, and Hooi, Bryan

Subjects
FOS: Computer and information sciences, Computer Science - Machine Learning, Computer Science - Computation and Language, Artificial Intelligence (cs.AI), Computer Science - Artificial Intelligence, Computation and Language (cs.CL), Information Retrieval (cs.IR), Computer Science - Information Retrieval, Machine Learning (cs.LG)
Abstract

Knowledge Extraction (KE), aiming to extract structural information from unstructured texts, often suffers from data scarcity and emerging unseen types, i.e., low-resource scenarios. Many neural approaches to low-resource KE have been widely investigated and achieved impressive performance. In this paper, we present a literature review towards KE in low-resource scenarios, and systematically categorize existing works into three paradigms: (1) exploiting higher-resource data, (2) exploiting stronger models, and (3) exploiting data and models together. In addition, we highlight promising applications and outline some potential directions for future research. We hope that our survey can help both the academic and industrial communities to better understand this field, inspire more ideas, and boost broader applications., Comment: Work in Progress. Github: \url{https://github.com/zjunlp/Low-resource-KEPapers}

Published
2022
Full Text
View/download PDF

109. Knowledge-Enhanced Collaborative Meta Learner for Long-Tail Recommendation

Author

Huajun Chen, Deng Shumin, and Bo Wen

Subjects
Information retrieval, Computer science, Order (business), Low resource, Prior probability, Collaborative filtering, Long tail, Recommender system, Representation (mathematics), Meaning (linguistics)
Abstract

Long-tail effect is common in recommender systems, meaning that a tiny number of users have lots of interaction with items, while the majority of users have extremely little interaction. Most of existing approaches to recommender systems, especially methods based on collaborative filtering, suffer severely from long-tail problems due to the low resource issue. In order to handle the problem of long-tail recommendation, we utilize knowledge graph to enrich the representation of users and items. As auxiliary prior information source, knowledge graph has been popular technology for recommender systems nowadays, while it is rarely exploited with respect to the low resource issue. In this paper, we propose a knowledge-enhanced collaborative meta learner, which combines the priors in knowledge graph and collaborative information between head users to promote long-tail recommendation performance. We conduct several experiments on two real world datasets for long-tail recommendation. The results show that our approach outperforms several commonly used recommendation methods in the long tail scenario.

Published
2021

131. Relationship between sexual sensation seeking and condom use among young men who have sex with men in China: testing a moderated mediation model.

Author

Wang, Wei, Yan, Hong, Duan, Zhizhou, Yang, Huiming, Li, Xiaoyan, Ding, Changmian, Deng, Shumin, and Li, Shiyue

Subjects
HIV infections, HUMAN sexuality, CROSS-sectional method, ATTITUDES toward sex, HEALTH literacy, SEX customs, FACTOR analysis, MEN who have sex with men, CONDOMS, DATA analysis software
Abstract

The present study incorporated a moderated mediation model to explore the role of attitude towards condom use in mediating the link between sexual sensation seeking (SSS) and condom use and whether this indirect link was modified by HIV-related knowledge among Chinese YMSM. Survey data were collected from a cross-sectional study conducted in Wuhan, China and 373 YMSM were recruited. The mediation and moderated mediation modelling analyses were performed with the software SPSS PROCESS macro. Mediation analysis indicated that attitude towards condom use partly mediated the link between SSS and condom use (indict effect = −0.158, P < 0.001). Moderation analysis found HIV-related knowledge acted as a moderator in the relationship between SSS and attitude towards condom use (interact effect = 0.089, P = 0.001). Final moderated mediation analysis demonstrated that the indirect effect from SSS to condom use through attitude towards condom use was moderated by HIV-related knowledge, that is the interaction between HIV-related knowledge and SSS was positively associated with attitude towards condom use (β = 0.101, P < 0.001). Therefore, increased YMSM-specific HIV-related knowledge education programs need to be conducted. Further longitudinal research is required to verify the findings of this study. [ABSTRACT FROM AUTHOR]

Published
2021
Full Text
View/download PDF

134. The temperature field simulation and energy saving design of ship borne asphalt tank

Author

Jihai Liu, Jinqing Liu, Jinchao Xiao, and Deng Shumin

Subjects
Pollution, business.industry, media_common.quotation_subject, Process (computing), Environmental pollution, Fuel oil, Field simulation, Asphalt, Environmental science, MATLAB, Process engineering, business, computer, Energy (signal processing), media_common, computer.programming_language
Abstract

Traditional asphalt tank has many defects in its structure, which affects the quality of asphalt finished product, so it is necessary to optimize its structure. This paper proposed a method to improve the problems existed in the heating process of traditional asphalt tank, such as uneven heating of asphalt, low heating efficiency, and environmental pollution. This method could simulate the temperature fields of the traditional asphalt tank based on MATLAB PDE toolbox, as well as compare and analyze the means and standard deviation of the two temperature fields. The experimental results show that this method could efficiently solve the problems about uneven heating of asphalt, low heating efficiency, and environmental pollution. Under the condition of enough asphalt liquid, the optimized asphalt tank could utilize lower temperature fuel oil to preserve heat, which finally achieves to save energy, reduce consumption and reduce pollution.

Published
2017

148. Cascaded Solution for Multi-domain Conditional Question Answering with Multiple-Span Answers

Author

Zhu, Junhao, Zhang, Min, Peng, Song, Yang, Hao, Qin, Ying, Zhang, Weidong, Han, Han, Ma, Miaomiao, Filipe, Joaquim, Editorial Board Member, Ghosh, Ashish, Editorial Board Member, Prates, Raquel Oliveira, Editorial Board Member, Zhou, Lizhu, Editorial Board Member, Zhang, Ningyu, editor, Wang, Meng, editor, Wu, Tianxing, editor, Hu, Wei, editor, and Deng, Shumin, editor

Published
2022
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results