Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks93 from public data
- 212
- 140
- Underwater-GAN: Underwater Image Restoration via Conditional Generative Adversarial Network107
This paper proposes Underwater-GAN, a conditional generative adversarial network for underwater image restoration, which uses Wasserstein GAN with gradient penalty term as the backbone network and designs the loss function as the sum of the loss of generative adversary network and the perceptual loss.
- 86
- 68
- Uncertainty-Driven Dehazing Network65
A novel uncertainty-driven dehazing network (UDN) is proposed that improves the dehazed results by exploiting the relationship between the uncertain and confident representations, and develops an uncertainty- driven self-distillation loss to improve the uncertain representation by transferring the knowledge from the confident one.
- AIM 2019 Challenge on Image Demoireing: Methods and Results57
This paper reviews the first-ever image demoireing challenge that was part of the Advances in Image Manipulation (AIM) workshop, held in conjunction with ICCV 2019, and focuses on the proposed solutions and their results.
- Supervised Hebb rule based feature selection for text classification54
Experimental results indicate that HRFS is effective to achieve better performance than the compared methods and also efficient because it can be described in the view of matrix operation to decrease complexity of feature selection.
- NTIRE 2020 Challenge on Image Demoireing: Methods and Results45
This paper reviews the Challenge on Image Demoireing that was part of the New Trends in Image Restoration and Enhancement (NTIRE) workshop, held in conjunction with CVPR 2020, and examines the current state-of-the-art in image and burst image demoireing problems.
- 42
- 41
- Research on customer opinion summarization using topic mining and deep neural network27
A novel framework for summarizing customer opinions from product reviews is proposed that identifies grammatically and semantically meaningful phrases which contain product attributes and their corresponding opinions from original product reviews by using grammar rules and the latent Dirichlet allocation (LDA) model.
- 25
- 25
- Online ad effectiveness evaluation with a two-stage method using a Gaussian filter and decision tree approach25
A two-stage method based on a Gaussian filter and a decision tree (M-GFDT) that helps in the removal of ineffective online ads as early as possible and achieves high accuracy in predicting effective online ads is proposed.
- Control of phytoplankton by oysters and the consequent impact on nitrogen cycling in a Subtropical Bay21
This study illustrates the influence of massive oysters on pelagic nitrogen cycling in an "oyster-remediated" estuary, providing implications for restoring oyster reefs to mitigate the symptoms of estuarine eutrophication.
- Distance Variance Score: An Efficient Feature Selection Method in Text Classification21
Experimental results indicate that DVS is able to select discriminative features and reduce the sparsity of DTM, which makes it much more efficient than LS.
- Filter feature selection methods for text classification: a review20
This paper aims to provide a comprehensive systematic review of existing filter feature selection methods for text classification by presenting a detailed discussion on mathematical designs, effectiveness and complexity of existing Filter Feature selection methods of different methodologies (super supervised methods, unsupervised methods and hybrid methods).
- 17
- Deep learning reconstruction for lumbar spine MRI acceleration: a prospective study16
Deep learning-reconstructed lumbar spine MRI protocol enabled a 45% reduction in scan time compared with conventional reconstruction, with comparable image quality and detection performance of common degenerative pathologies.
- ERNIE 5.0 Technical Report15
ERNIE 5.0 represents the first production-scale realization of a trillion-parameter unified autoregressive model that supports both multimodal understanding and generation and systematically addresses the challenges of scaling reinforcement learning to unified foundation models.
- Research on BP Neural Network Recommendation Model Fusing User Reviews and Ratings14
Experimental results indicate that the outstanding performance of NNFR model than comparative recommendation algorithms on rating predictions and top-k recommendations can still produce high-quality recommendation results in the scenarios of sparse data.
- 14
- 13
- 12
- 11
- [Responses of Cynodon dactylon population in hydro-fluctuation belt of Three Gorges Reservoir area to flooding-drying habitat change].11
The unique adaption strategies of C. dactylon to the flooding-drying habitat change in the shallow water section were the accelerated elongation growth and the increased stem biomass allocation, while those in the deep water sections were the increased node number of primary and secondary branches, increased number of the branches, and increased leaf biomass allocation.
- Toward Better Understanding Older Adults: A Biography Brief Timeline Extraction Approach10
AlberT Based Text Extraction Network (ABTE-NET) is proposed to generate valuable event timelines to address the problem of redundant and disordered life stories in older adults, and these timelines are better organized and more concise than original life stories.
- Investigating the economic returns of mobile heat storage devices in the multi-stage closed-loop supply chain10
A numerical model is proposed to investigate the dynamic supply chain of the M-WHR and suggested that by optimizing the production plan based on the market demand, the supply chain can reach the Pareto state and minimize its production cost and idle production recourses.
- Multi-sensor data fusion based on consistency test and sliding window variance weighted algorithm in sensor networks9
Under the condition that the abnormal sensor data has been amended or removed, the proposed algorithm has better performances on precision compared with other existing algorithms.
- [Life form spectra, leaf character, and hierarchical-synusia structure of vascular plants in Thuja sutchuehensis community].8
The T. sutchuenensis community had three sub-layers, i.e., tree layer, shrub layer, and herb layer, with lesser interlayer plants, respectively composed by phanemphyte evergreen coniferophyte, broadleaf and deciduous broad-leaf plants, as well as hemicryptophytes, geophyt, and annual plants.
- Comparison on Feature Selection Methods for Text Classification7
This paper introduces and discusses a series of typical feature selection methods which are commonly involved in previous studies for text classification, and conducts comparison experiments on four benchmark datasets to compare the effectiveness of twenty typical features selection methods.
- 7
- 6
- 5
- Utility-based feature selection for text classification5
This work represents the successful research on the leveraging economics theory for developing a semantically rich feature selection method to improve text classification and provides an economics explanation of term weighting for information retrieval and semantic information acquisition in textual documents.
- 5
- Simulation of Crop Evaportranspiration Based on BP Neural Network Model and Grey Relational Analysis5
- A Water Quality Monitoring Method Based on Fuzzy Comprehensive Evaluation in Wireless Sensor Networks5
Simulation result shows that the novel water quality monitoring method named WQMMFCE has a great advantage over the centralized one on energy saving and can prolong the lifetime of the network.
- 5
- Identification of cell surface markers for acute myeloid leukemia prognosis based on multi-model analysis4
A 9-CSMs prognostic model was constructed that served as an independent prognostic factor for the survival of AML patients and held the potential for guiding drug therapy.
- Research on the correlation between the degree of digitization of manufacturing process links and economic benefits4
The relationship between digitalization of manufacturing process links and economic benefits can provide theoretical and empirical support for the digital transformation of China's manufacturing industry and high-quality development of economy.
- 4
- Feature selection based on long short term memory for text classification4
A deep learning based feature selection method that extends the limitation of term frequency information by applying deep features for feature selection and demonstrates better performance in selecting discriminative terms than comparative methods.
- 4
- 4
- 4
- 3
- 3
- A Feature Selection Method Based on a Convolutional Neural Network for Text Classification3
A new ranking feature selection method using a pre-trained three-layer CNN to capture deep features of terms and selects discriminative terms according to the combination of deep features and term frequencies, aiming to improve performance of feature selection for text classification.
- The Development Status of the Manufacturing Industry and the Impact of Digital Characteristics from the Perspective of Innovation3
In general, manufacturing digital transformation in China highly relates to regional economy development and the development speed of the two systems tends to be stable from 2017 to 2021, so the regions with different coordinated development types should formulate corresponding policies to accelerate manufacturing digital transformation and regional economy development.
- A 2020 perspective on “Online ad effectiveness evaluation with a two-stage method using a Gaussian filter and decision tree approach”3
A two-stage method based on a Gaussian filter and a decision tree (M-GFDT) that achieves high accuracy in predicting effective online ads and aids in the removal of ineffective online ads as early as possible is proposed.
- 3
- Neural-Network-Based Outcome Classification for Nursing Care2
A machine learning model is built to predict the outcome domain only based on outcome indictors, so that when the nurses obtain the indictors after the evaluation, the model could output the predicted domain to guide the nursing care plan formulation.
- 2
- Salt Transfer Law for Cotton Field with Drip Irrigation Under The Plastic Mulch in Arid Zone2
The results show that with the retardation of postemergences, the accumulation of salt in 0-50cm soil layers decrease in different extent and the accumulation in horizontal direction at the narrow row is more than any other position and the one under the emitter is least.
- Hydroxylation and Biological Activity Evaluation of Isosteviol C12 by P450 BM31
Compound similarity analysis, molecular docking, and site‐directed mutagenesis are utilized to successfully achieve the regioselective hydroxylation of isosteviol at the 12β position, yielding 12β‐hydroxyisosteviol, catalyzed by P450 BM3, which exhibits superior anticancer and anti‐inflammatory properties relative to its precursor, isosteviol.
- 1
- Research on Cross-Platform Image Recommendation Model Fusing Text Information1
A cross-platform image recommendation model (FITIFCIR), which fuses text and image information to achieve cross-platform data recommendation and indicates that the FITIFCIR outperforms baseline models.
- 1
- 1
- 1
- 1
- Design of Portable ECG Measurement Device1
A new portable ECG measurable device with high processing is introduced which is small and lower power and can detect QRS complex and analyze in real-time and embed file system, so the ECG data can be stored in text form and data's readability and translatability are enhanced.
- 1
- Deep Learning-based Method for Expressing Knowledge Boundary of Black-Box LLM–
LSCL (LLM-Supervised Confidence Learning), a deep learning-based method for expressing the knowledge boundaries of black-box LLMs, is proposed based on the knowledge distillation framework, and significantly outperforms existing baseline models on metrics such as accuracy and recall rate.
- Ad Keyframe Extraction and Product Detection Framework in Short Video Advertising–
An Advertising Keyframe Extraction and Product Detection Framework (AKEPDF) is proposed, which operates in two stages: first, using an unsupervised K‐Nearest Neighbor algorithm to segment videos and select clear frames as keyframes; then, using EfficientViT‐YOLOv8n for product detection in keyframes.
- –
- SIM Card Delivery Time Prediction Based on the Interpretable NSGA‐III‐XGBoost–
Predicting the delivery time of SIM cards in the case of first delivery failure provides couriers with a reference for the final delivery time, helping them make reasonable arrangements within the time slots scheduled with consumers and thereby enhancing delivery efficiency.
- Online Shopping Service Optimization Based on Topic Mining and Process Chain Network–
This paper presents a data‐driven methodology for service quality management of online shopping platforms and optimization of the purchase processes of online shopping platforms based on the discovered service problems and PCN.
- –
- –
- –
- –
- –
- –
- –
- Multi-Sensor Aggregation for Path Reliability Estimation in WSNs–
In wireless sensor networks, data transmission with energy-effective path faces the problems of lower reliability and link failures, so a redundant transmission approach is employed to improve path reliability and to balance the network energy consumption.
- –
- –
- –
- –
- –
- Solution strategies for a supply chain deterministic model–
A mixed integer-programming model of 3-tier supply chain with multi-plant, multi-warehouse, and multi-retailer, multiperiod and restricted capacity is built that aims to evaluate and optimize the overall performance of the supply chain.
- –
- –
- –
- –
- –
- The study of a new type of intellectualized washing stomach machine–
The authors discuss the design of a new type of intellectualized washing stomach machine that consists of sensors, control circuit, CPU, mini pump, touching panel and display panel to realize automation and make users use it conveniently.
- LCDfor the portablemyocardialinfractionmonitor–
The LCD monitor can provide real-time, accurate, high-resolution display of ECG waveforms as well as characters and has been introduced to fulfill portable myocardial infraction monitoring system with the functions of auto-diagnosing, alarming and administration.
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.