Lingling Li

Lingling Li


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 November 2022 in Algorithms
Verified
Journal Article
Open access
PDF file
JLJiamu Li
Jiamu Li
JZJi Zhang
Ji Zhang
MBMohamed Jaward Bah
Mohamed Jaward Bah
JWJian WANG
Jian WANG
YZYouwen Zhu
Youwen Zhu
KZKexin Zhang
Kexin Zhang
When dealing with high-dimensional data, such as in biometric, e-commerce, or industrial applications, it is extremely hard to capture the abnormalities in full space due to the curse of dimensionality. Furthermore, it is becoming increasingly complicated but essential to provide interpretations for outlier detection results in high-dimensional space as a consequence of the large number of features. To alleviate these issues, we propose a new model based on a Variational AutoEncoder and Genetic Algorithm (VAEGA) for detecting outliers in subspaces of high-dimensional data. The proposed model employs a neural network to create a probabilistic dimensionality reduction variational autoencoder (VAE) that applies its low-dimensional hidden space to characterize the high-dimensional inputs. Then, the hidden vector is sampled randomly from the hidden space to reconstruct the data so that it closely matches the input data. The reconstruction error is then computed to determine an outlier score, and samples exceeding the threshold are tentatively identified as outliers. In the second step, a genetic algorithm (GA) is used as a basis for examining and analyzing the abnormal subspace of the outlier set obtained by the VAE layer. After encoding the outlier dataset’s subspaces, the degree of anomaly for the detected subspaces is calculated using the redefined fitness function. Finally, the abnormal subspace is calculated for the detected point by selecting the subspace with the highest degree of anomaly. The clustering of abnormal subspaces helps filter outliers that are mislabeled (false positives), and the VAE layer adjusts the network weights based on the false positives. When compared to other methods using five public datasets, the VAEGA outlier detection model results are highly interpretable and outperform or have competitive performance compared to current contemporary methods.

28 January 2024 in Computer Networks
Verified
Journal Article
Closed access
XYXiaoyun Yuan
Xiaoyun Yuan
TLTengyao Li
Tengyao Li
LLLingling Li
Lingling Li
RLRuixiang Li
Ruixiang Li
ZWZhiquan Wang
Zhiquan Wang
LXLuo Xiangyang
Luo Xiangyang

14 June 2024 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
MYMing Yin
Ming Yin
XLXin Liu
Xin Liu
JGJunli Gao
Junli Gao
HYHaoliang Yuan
Haoliang Yuan
TJTaisong Jin
Taisong Jin
LLLingling Li
Lingling Li
Multiview representation learning has recently attracted significant attention in the machine learning and computer vision community. However, during fusing information from multiple views, existing work often neglect to exploit the complementary information in each view and endow with the interpretability of model. To this end, in this article, we propose a multiview attention fusion information bottleneck network, termed by MVAIBNet. Specifically, MVAIBNet deliberately develops dual-path reconstructions to extract latent embeddings of views, where view-peculiar representations are disentangled from the embeddings using $\beta$ -VAE, to help reconstruct each view. Then, to align and fuse view-common representations, an attention fusion unit, namely the multiview channel fusion unit (MVCFU), is presented accordingly. Furthermore, relying on the information bottleneck principle, we integrate the consistency information and specificity information of the views to prompt a compact semantic representation of multiple views with balancing the complementarity and consistency among multiple views flexibly. Extensive experimental results on four real-world datasets show that our algorithm achieves encouraging performance on several evaluation metrics compared to the state-of-the-art methods.

30 September 2024 in Expert Systems with Applications
Verified
Journal Article
Closed access
HWHaiyan Wang
Haiyan Wang
YZYanjie Zhou
Yanjie Zhou
XNXiangdong Niu
Xiangdong Niu
DLDaijun Liu
Daijun Liu
LLLingling Li
Lingling Li
ZWZongmin Wang
Zongmin Wang

20 March 2026 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
ZZZhuangzhuang Zhou
Zhuangzhuang Zhou
XZXuewu Zhao
Xuewu Zhao
XZXuezhuan Zhao
Xuezhuan Zhao
YZYingying Zhu
Yingying Zhu
ZZZunjin Zhao
Zunjin Zhao
LLLingling Li
Lingling Li
Optimal efficiency is always the primary pursuit of the remote sensing community. However, the blurred appearance in small objects causes increasingly deeper and wider detectors, posing a significant challenge for deployment. Knowledge Distillation (KD), which can shorten the time from image acquisition to interpretation, has become a hotspot. In this paper, contrary to traditional feature imitation methods, we propose generative-driven knowledge distillation for small object detection (GenDist), which reconstructs teacher-like representations from the partially masked student features to achieve knowledge transfer. Masking during training strengthens the student’s ability to generalize in complex scenes. Specifically, GenDist contains three parts: Online Self-Correct Mask (O-SCM), Dynamic Activation Global Context (DAGC) and Sampling-Aligned Generation (SAG). First, O-SCM is utilized to identify informative masking regions that guide the student to reconstruct the teacher-like features. By integrating an online self-correction mechanism, it adaptively directs generative supervision toward regions that are ignored or misinterpreted by students. Then, DAGC refines teacher features via global context modeling and critical feature selection, facilitating more accurate object detail capture by the student model. Finally, SAG improves the quality of feature reconstruction by introducing the learnable offset sampling, enabling students to learn irregular areas such as the small instances’ edges and texture transitions. Our proposed method finds the sweet spot between precision and speed in an elegant way. GenDist yields stable performance gains for student models across the DIOR, AI-TOD, SODA-D, and SSDD datasets. For the lightweight detectors, it cuts the model’s parameter count by roughly 38.2%, while keeping accuracy loss under 3.9%.

29 November 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
GWGuo Wei
Guo Wei
SDShichang Ding
Shichang Ding
HYHaifeng Yang
Haifeng Yang
WLWenyan Liu
Wenyan Liu
MYMeijuan Yin
Meijuan Yin
LLLingling Li
Lingling Li
Recently, some criminals have begun to use multimedia steganography to conduct covert communication, such as transmitting stolen trade secrets. After using steganalysis to find covert communication entities, obtaining their locations can effectively help criminal forensics. This paper proposes a novel localization method of wireless covert communication entity for post-steganalysis. The method is based on hybrid particle swarm optimization and gray wolf optimization to improve localization precision (ILP-PSOGWO). In this method, firstly, the relationship model between received signal strength (RSS) and distance is constructed for the indoor environment where the target node exists. Secondly, dichotomy is used to narrow the region where the target node is located. Then, the weighted distance strategy is used to select the reference point locations with strong and stable RSS. Finally, the intersection region of the reference points is taken as the region where the target node is located, and the hybrid PSOGWO is used to locate and optimize the target node location. Experimental results demonstrate that ILP-PSOGWO can maintain high stability, and 90% of the localization errors are lower than 0.9012 m. In addition, compared with the existing methods of PSO, GWO and extended weighted centroid localization (EWCL), the average localization error of ILP-PSOGWO is also reduced by 28.2–49.0%.