Ruijing Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Ruijing Yang has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Ziyu Guan
School of Computer Science and Technology, Xidian University, Xi’an, China
2 shared publications - Cai Xu
School of Computer Science and Technology, Xidian University, China
1 shared publication - Fei Wang
Xi’an Jiaotong University
1 shared publication - Fei Xie
Xidian University
1 shared publication - Guoying Zhao
University of Oulu, Oulu, Finland
1 shared publication Jinye Peng
Northwest University
1 shared publication
Latest Publications
Assessing weak interaction and threshold effects of the trophic field overlap index on the relationship between species uniqueness and centrality in food webs
In addition to considering the influence of a species on other species (centrality), the identification of keystone species should also consider whether it has a unique network position (uniqueness). Clarifying the correlation between species centrality and uniqueness, or whether central species exhibit positional redundancy in food webs, can enhance the understanding of mechanisms stabilizing food webs. However, the findings from analyses utilizing different measures of uniqueness are contentious. In this study, we constructed the food web of the eutrophic Zhangze Lake (northern China) using stable isotopes to estimate the contribution of food sources to consumers. We then calculated trophic field overlap index to assess the uniqueness of species using this quantified food web information. Additionally, we analyzed how the weak interaction and the threshold value under which interactions are considered as weak influence the relationship between species uniqueness and centrality. The findings revealed that, in contrast to the traditional trophic field overlap index, the improved index identified the basal species with the strongest trophic interactions with other species and the filter‐feeding fishes with the fewest interactors as the most unique species, while the basal species were considered the least unique under the traditional index. The ranking results of the traditional index and the centrality showed a significant negative correlation, whereas the improved index displayed a positive correlation, which was caused by the comprehensive consideration of threshold and weak interaction. In addition, based on the data analysis, the suggestion was given for the selection of indices used to identify unique species, and whether central species have positional redundancy was explored. The methodology used in study provides a practical solution for the identification of the most unique species, points out and solves the problems faced in the uniqueness analysis, and advances the understanding of the mechanisms underlying food web robustness.
Assessing weak interaction and threshold effects of the trophic field overlap index on the relationship between species uniqueness and centrality in food webs
In addition to considering the influence of a species on other species (centrality), the identification of keystone species should also consider whether it has a unique network position (uniqueness). Clarifying the correlation between species centrality and uniqueness, or whether central species exhibit positional redundancy in food webs, can enhance the understanding of mechanisms stabilizing food webs. However, the findings from analyses utilizing different measures of uniqueness are contentious. In this study, we constructed the food web of the eutrophic Zhangze Lake (northern China) using stable isotopes to estimate the contribution of food sources to consumers. We then calculated trophic field overlap index to assess the uniqueness of species using this quantified food web information. Additionally, we analyzed how the weak interaction and the threshold value under which interactions are considered as weak influence the relationship between species uniqueness and centrality. The findings revealed that, in contrast to the traditional trophic field overlap index, the improved index identified the basal species with the strongest trophic interactions with other species and the filter‐feeding fishes with the fewest interactors as the most unique species, while the basal species were considered the least unique under the traditional index. The ranking results of the traditional index and the centrality showed a significant negative correlation, whereas the improved index displayed a positive correlation, which was caused by the comprehensive consideration of threshold and weak interaction. In addition, based on the data analysis, the suggestion was given for the selection of indices used to identify unique species, and whether central species have positional redundancy was explored. The methodology used in study provides a practical solution for the identification of the most unique species, points out and solves the problems faced in the uniqueness analysis, and advances the understanding of the mechanisms underlying food web robustness.
- 5 August 2022
- Published by Springer Nature in International Journal of Machine Learning and Cybernetics
- Vol. 13 (12) , 3727-3742
Representation learning from noisy user-tagged data for sentiment classification
- 5 August 2022
- Published by Springer Nature in International Journal of Machine Learning and Cybernetics
- Vol. 13 (12) , 3727-3742
Representation learning from noisy user-tagged data for sentiment classification
- Open Access
- 1 August 2021
- Published by International Joint Conferences on Artificial Intelligence
- Vol. 2, 1231-1237
Non-contact Pain Recognition from Video Sequences with Remote Physiological Measurements Prediction
Automatic pain recognition is paramount for medical diagnosis and treatment. The existing works fall into three categories: assessing facial appearance changes, exploiting physiological cues, or fusing them in a multi-modal manner. However, (1) appearance changes are easily affected by subjective factors which impedes objective pain recognition. Besides, the appearance-based approaches ignore long-range spatial-temporal dependencies that are important for modeling expressions over time; (2) the physiological cues are obtained by attaching sensors on human body, which is inconvenient and uncomfortable. In this paper, we present a novel multi-task learning framework which encodes both appearance changes and physiological cues in a non-contact manner for pain recognition. The framework is able to capture both local and long-range dependencies via the proposed attention mechanism for the learned appearance representations, which are further enriched by temporally attended physiological cues (remote photoplethysmography, rPPG) that are recovered from videos in the auxiliary task. This framework is dubbed rPPG-enriched Spatio-Temporal Attention Network (rSTAN) and allows us to establish the state-of-the-art performance of non-contact pain recognition on publicly available pain databases. It demonstrates that rPPG predictions can be used as an auxiliary task to facilitate non-contact automatic pain recognition.
- 1 August 2021
- Published by International Joint Conferences on Artificial Intelligence
- Vol. 2, 1231-1237
Open AccessNon-contact Pain Recognition from Video Sequences with Remote Physiological Measurements Prediction
Automatic pain recognition is paramount for medical diagnosis and treatment. The existing works fall into three categories: assessing facial appearance changes, exploiting physiological cues, or fusing them in a multi-modal manner. However, (1) appearance changes are easily affected by subjective factors which impedes objective pain recognition. Besides, the appearance-based approaches ignore long-range spatial-temporal dependencies that are important for modeling expressions over time; (2) the physiological cues are obtained by attaching sensors on human body, which is inconvenient and uncomfortable. In this paper, we present a novel multi-task learning framework which encodes both appearance changes and physiological cues in a non-contact manner for pain recognition. The framework is able to capture both local and long-range dependencies via the proposed attention mechanism for the learned appearance representations, which are further enriched by temporally attended physiological cues (remote photoplethysmography, rPPG) that are recovered from videos in the auxiliary task. This framework is dubbed rPPG-enriched Spatio-Temporal Attention Network (rSTAN) and allows us to establish the state-of-the-art performance of non-contact pain recognition on publicly available pain databases. It demonstrates that rPPG predictions can be used as an auxiliary task to facilitate non-contact automatic pain recognition.