Huiwen Zheng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Huiwen Zheng has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Taigang Liu
College of Information Technology, Shanghai Ocean University, Shanghai, 201306, China
2 shared publications - Wanyue Xiao
School of Information, Syracuse University, Syracuse, NY 13244, USA
2 shared publications - Yang Yang
School of Information Management, Nanjing University, Nanjing 210023, China
2 shared publications - Changhao Zhou
GDS Holdings, Shanghai, China
1 shared publication - Ding Li
Key Laboratory of High-Confidence Software Technologies (MOE), School of Computer Science, Peking University, Beijing, China
1 shared publication - Guocong Lin
Tsinghua University, Institute for AI Industry Research (AIR), Beijing, China
1 shared publication
Latest Publications
- Open Access
- 7 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 3908-3919
Towards Reliable Alarm Flood Reduction in Data Center via Grammar-Guided Verifiable LLM Reasoning
- 7 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 3908-3919
Open AccessTowards Reliable Alarm Flood Reduction in Data Center via Grammar-Guided Verifiable LLM Reasoning
- Open Access
- 21 April 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Internet of Things
- Vol. 7 (2) , 1-26
ChatDC: Geometric-aware Data Center Digital Twin Generation via Large Language Models
A modern data center is supported by an Internet of Assets (IoA), a specialized Internet of Things (IoT), where the sim-ready assets encompass physical properties and connected data streams for passive data collection and proactive simulation. A digital twin (DT) is a virtual replica of the IoA system with a proper level of abstraction, integrating asset-level dynamics models to simulate system-wide behavior, prototype designs, and conduct what-if analysis. However, creating high-fidelity DTs is hindered by the manual effort needed to encode complex geometric layouts and domain-specific design constraints. While large language models (LLMs) offer automation potential, existing methods struggle to generate geometrically plausible and functionally valid DT scenes due to limited domain integration. To bridge the gaps, we propose ChatDC, a conversational system that leverages LLMs to automate data center digital twin generation through a S egment- G enerate- O ptimize (SGO) workflow. ChatDC integrates domain knowledge via a dedicated code library named DCBuild and employs SGO to decompose user prompts, generate initial structures, and optimize layouts in compliance with data center design constraints. Evaluation shows that ChatDC outperforms other baselines with 98% success rate on scratch generation tasks and reduces the average makespan by 10×. Ablation study reveals that the SGO design increases the generation success rate by 65% at most. Furthermore, computational fluid dynamics simulations validate the physical plausibility, confirming the readiness of generated DTs for real-world analysis.
- 21 April 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Internet of Things
- Vol. 7 (2) , 1-26
Open AccessChatDC: Geometric-aware Data Center Digital Twin Generation via Large Language Models
A modern data center is supported by an Internet of Assets (IoA), a specialized Internet of Things (IoT), where the sim-ready assets encompass physical properties and connected data streams for passive data collection and proactive simulation. A digital twin (DT) is a virtual replica of the IoA system with a proper level of abstraction, integrating asset-level dynamics models to simulate system-wide behavior, prototype designs, and conduct what-if analysis. However, creating high-fidelity DTs is hindered by the manual effort needed to encode complex geometric layouts and domain-specific design constraints. While large language models (LLMs) offer automation potential, existing methods struggle to generate geometrically plausible and functionally valid DT scenes due to limited domain integration. To bridge the gaps, we propose ChatDC, a conversational system that leverages LLMs to automate data center digital twin generation through a S egment- G enerate- O ptimize (SGO) workflow. ChatDC integrates domain knowledge via a dedicated code library named DCBuild and employs SGO to decompose user prompts, generate initial structures, and optimize layouts in compliance with data center design constraints. Evaluation shows that ChatDC outperforms other baselines with 98% success rate on scratch generation tasks and reduces the average makespan by 10×. Ablation study reveals that the SGO design increases the generation success rate by 65% at most. Furthermore, computational fluid dynamics simulations validate the physical plausibility, confirming the readiness of generated DTs for real-world analysis.
- Open Access
- 18 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 1786-1790
ADL: Adaptive Distribution Learning Framework for Multi-Scenario CTR Prediction
Large-scale commercial platforms usually involve numerous business scenarios for diverse business strategies. To provide click-through rate (CTR) predictions for multiple scenarios simultaneously, existing promising multi-scenario models explicitly construct scenario-specific networks by manually grouping scenarios based on particular business strategies. Nonetheless, this pre-defined data partitioning process heavily relies on prior knowledge, and it may neglect the underlying data distribution of each scenario, hence limiting the model's representation capability. Regarding the above issues, we propose Adaptive Distribution Learning (ADL): an end-to-end optimization distribution framework which is composed of a clustering process and classification process. Specifically, we design a distribution adaptation module with a customized dynamic routing mechanism. Instead of introducing prior knowledge for pre-defined data allocation, this routing algorithm adaptively provides a distribution coefficient for each sample to determine which cluster it belongs to. Each cluster corresponds to a particular distribution so that the model can sufficiently capture the commonalities and distinctions between these distinct clusters. Our results on both public and large-scale industrial datasets show the effectiveness and efficiency of ADL: the model yields impressive prediction accuracy with more than 50% reduction in time cost during the training phase when compared to other methods.
- 18 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 1786-1790
Open AccessADL: Adaptive Distribution Learning Framework for Multi-Scenario CTR Prediction
Large-scale commercial platforms usually involve numerous business scenarios for diverse business strategies. To provide click-through rate (CTR) predictions for multiple scenarios simultaneously, existing promising multi-scenario models explicitly construct scenario-specific networks by manually grouping scenarios based on particular business strategies. Nonetheless, this pre-defined data partitioning process heavily relies on prior knowledge, and it may neglect the underlying data distribution of each scenario, hence limiting the model's representation capability. Regarding the above issues, we propose Adaptive Distribution Learning (ADL): an end-to-end optimization distribution framework which is composed of a clustering process and classification process. Specifically, we design a distribution adaptation module with a customized dynamic routing mechanism. Instead of introducing prior knowledge for pre-defined data allocation, this routing algorithm adaptively provides a distribution coefficient for each sample to determine which cluster it belongs to. Each cluster corresponds to a particular distribution so that the model can sufficiently capture the commonalities and distinctions between these distinct clusters. Our results on both public and large-scale industrial datasets show the effectiveness and efficiency of ADL: the model yields impressive prediction accuracy with more than 50% reduction in time cost during the training phase when compared to other methods.
- Open Access
- 26 February 2021
- Published by Association for Computing Machinery (ACM)
- p. 296-301
A binary variational mode decomposition algorithm based on similarity matching
The existing algorithms in signal processing had the problems of incomplete separation and inability to adaptively decompose signals. The VMD algorithm can be used to decompose the composite signal, but the number of IMFs needs to be artificially set before decomposition. This paper proposes a binary variational mode decomposition algorithm based on similarity matching, which combines the tree decomposition with the similarity measurement to separate the other modes mixed in the current mode signal, and finally obtains the pure sub-signal through superposition. According to the sub-signal mutual information between the signal and the original signal removes noise components. Simulation analysis shows that BVMD has a good performance of denoising and signal separation.
- 26 February 2021
- Published by Association for Computing Machinery (ACM)
- p. 296-301
Open AccessA binary variational mode decomposition algorithm based on similarity matching
The existing algorithms in signal processing had the problems of incomplete separation and inability to adaptively decompose signals. The VMD algorithm can be used to decompose the composite signal, but the number of IMFs needs to be artificially set before decomposition. This paper proposes a binary variational mode decomposition algorithm based on similarity matching, which combines the tree decomposition with the similarity measurement to separate the other modes mixed in the current mode signal, and finally obtains the pure sub-signal through superposition. According to the sub-signal mutual information between the signal and the original signal removes noise components. Simulation analysis shows that BVMD has a good performance of denoising and signal separation.
- 1 May 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 18, 1-6
RSS-based Indoor Passive Localization Using Clustering and Filtering in a LTE Network
Nowadays, fingerprint positioning is the mainstream method in indoor positioning and Weighted-K-Nearest-Neighbor (WKNN) is most widely used in fingerprint matching. However, the fingerprints which are far away from each other might also be similar and the target fingerprints have to match with all fingerprints every time, which results in unsatisfactory accuracy and efficiency of WKNN. In this paper, we propose an algorithm to classify regions with comprehensive consideration of geographic location and fingerprint similarity, meanwhile preprocessing and filtering the data are used to improve the accuracy. In addition, we propose that the cellular signal in Long Term Evolution (LTE) is more suitable to be the signal source in indoor passive localization scenario. Results show that in the LTE environment, our proposed algorithm effectively reduces the positioning error by about 24% and improves the convergence speed compared with WKNN.
- 1 May 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 18, 1-6
RSS-based Indoor Passive Localization Using Clustering and Filtering in a LTE Network
Nowadays, fingerprint positioning is the mainstream method in indoor positioning and Weighted-K-Nearest-Neighbor (WKNN) is most widely used in fingerprint matching. However, the fingerprints which are far away from each other might also be similar and the target fingerprints have to match with all fingerprints every time, which results in unsatisfactory accuracy and efficiency of WKNN. In this paper, we propose an algorithm to classify regions with comprehensive consideration of geographic location and fingerprint similarity, meanwhile preprocessing and filtering the data are used to improve the accuracy. In addition, we propose that the cellular signal in Long Term Evolution (LTE) is more suitable to be the signal source in indoor passive localization scenario. Results show that in the LTE environment, our proposed algorithm effectively reduces the positioning error by about 24% and improves the convergence speed compared with WKNN.
- Open Access
- 28 March 2020
- Published by Wiley in Computational and Mathematical Methods in Medicine
- Vol. 2020, 1-10
HMMPred: Accurate Prediction of DNA-Binding Proteins Based on HMM Profiles and XGBoost Feature Selection
Prediction of DNA-binding proteins (DBPs) has become a popular research topic in protein science due to its crucial role in all aspects of biological activities. Even though considerable efforts have been devoted to developing powerful computational methods to solve this problem, it is still a challenging task in the field of bioinformatics. A hidden Markov model (HMM) profile has been proved to provide important clues for improving the prediction performance of DBPs. In this paper, we propose a method, called HMMPred, which extracts the features of amino acid composition and auto- and cross-covariance transformation from the HMM profiles, to help train a machine learning model for identification of DBPs. Then, a feature selection technique is performed based on the extreme gradient boosting (XGBoost) algorithm. Finally, the selected optimal features are fed into a support vector machine (SVM) classifier to predict DBPs. The experimental results tested on two benchmark datasets show that the proposed method is superior to most of the existing methods and could serve as an alternative tool to identify DBPs.
- 28 March 2020
- Published by Wiley in Computational and Mathematical Methods in Medicine
- Vol. 2020, 1-10
Open AccessHMMPred: Accurate Prediction of DNA-Binding Proteins Based on HMM Profiles and XGBoost Feature Selection
Prediction of DNA-binding proteins (DBPs) has become a popular research topic in protein science due to its crucial role in all aspects of biological activities. Even though considerable efforts have been devoted to developing powerful computational methods to solve this problem, it is still a challenging task in the field of bioinformatics. A hidden Markov model (HMM) profile has been proved to provide important clues for improving the prediction performance of DBPs. In this paper, we propose a method, called HMMPred, which extracts the features of amino acid composition and auto- and cross-covariance transformation from the HMM profiles, to help train a machine learning model for identification of DBPs. Then, a feature selection technique is performed based on the extreme gradient boosting (XGBoost) algorithm. Finally, the selected optimal features are fed into a support vector machine (SVM) classifier to predict DBPs. The experimental results tested on two benchmark datasets show that the proposed method is superior to most of the existing methods and could serve as an alternative tool to identify DBPs.
- Open Access
- 1 January 2020
- Published by Wiley in BioMed Research International
- Vol. 2020 (1) , 7297631
PredDBP‐Stack: Prediction of DNA‐Binding Proteins from HMM Profiles using a Stacked Ensemble Method
DNA‐binding proteins (DBPs) play vital roles in all aspects of genetic activities. However, the identification of DBPs by using wet‐lab experimental approaches is often time‐consuming and laborious. In this study, we develop a novel computational method, called PredDBP‐Stack, to predict DBPs solely based on protein sequences. First, amino acid composition (AAC) and transition probability composition (TPC) extracted from the hidden markov model (HMM) profile are adopted to represent a protein. Next, we establish a stacked ensemble model to identify DBPs, which involves two stages of learning. In the first stage, the four base classifiers are trained with the features of HMM‐based compositions. In the second stage, the prediction probabilities of these base classifiers are used as inputs to the meta‐classifier to perform the final prediction of DBPs. Based on the PDB1075 benchmark dataset, we conduct a jackknife cross validation with the proposed PredDBP‐Stack predictor and obtain a balanced sensitivity and specificity of 92.47% and 92.36%, respectively. This outcome outperforms most of the existing classifiers. Furthermore, our method also achieves superior performance and model robustness on the PDB186 independent dataset. This demonstrates that the PredDBP‐Stack is an effective classifier for accurately identifying DBPs based on protein sequence information alone.
- 1 January 2020
- Published by Wiley in BioMed Research International
- Vol. 2020 (1) , 7297631
Open AccessPredDBP‐Stack: Prediction of DNA‐Binding Proteins from HMM Profiles using a Stacked Ensemble Method
DNA‐binding proteins (DBPs) play vital roles in all aspects of genetic activities. However, the identification of DBPs by using wet‐lab experimental approaches is often time‐consuming and laborious. In this study, we develop a novel computational method, called PredDBP‐Stack, to predict DBPs solely based on protein sequences. First, amino acid composition (AAC) and transition probability composition (TPC) extracted from the hidden markov model (HMM) profile are adopted to represent a protein. Next, we establish a stacked ensemble model to identify DBPs, which involves two stages of learning. In the first stage, the four base classifiers are trained with the features of HMM‐based compositions. In the second stage, the prediction probabilities of these base classifiers are used as inputs to the meta‐classifier to perform the final prediction of DBPs. Based on the PDB1075 benchmark dataset, we conduct a jackknife cross validation with the proposed PredDBP‐Stack predictor and obtain a balanced sensitivity and specificity of 92.47% and 92.36%, respectively. This outcome outperforms most of the existing classifiers. Furthermore, our method also achieves superior performance and model robustness on the PDB186 independent dataset. This demonstrates that the PredDBP‐Stack is an effective classifier for accurately identifying DBPs based on protein sequence information alone.