Deep Clustering Method Research Articles

A clustering ensemble algorithm for handling deep embeddings using cluster confidence

Abstract Clustering ensemble, which aims to learn a robust consensus clustering from multiple weak base clusterings, has achieved promising performance on various applications. With the development of big data, the scale and complexity of data is constantly increasing. However, most existing clustering ensemble methods typically employ shallow clustering algorithms to generate base clusterings. When confronted with high-dimensional complex data, these shallow algorithms fail to fully utilize the intricate features present in the latent data space. As a result, the quality and diversity of the generated base clusterings are insufficient, thus affecting the subsequent ensemble performance. To address this issue, we propose a novel clustering ensemble algorithm for handling deep embeddings using cluster confidence (CEDECC) to improve the robustness and performance. Instead of simply combining deep clustering with clustering ensembles, we take into consideration that the performance of existing deep clustering methods heavily relies on the quality of low-dimensional embeddings generated during the pre-training stage. The quality of embeddings is unstable due to the influence of different initialization parameters. In CEDECC, specifically, we first construct a cluster confidence measure to evaluate the quality of low-dimensional embeddings. Typically, high-quality low-dimensional embeddings yield accurate clustering results with the same model parameters. Then, we utilize multiple high-quality embeddings to generate the base partitions. In the ensemble strategy phase, we consider the cluster-wise diversity and propose a novel ensemble cluster estimation to improve the overall consensus performance of the model. Extensive experiments on three benchmark datasets and four real-world biological datasets have demonstrated that the proposed CEDECC consistently outperforms the state-of-the-art clustering ensemble methods.

Graph convolutional network for water network partitioning

The partitioning of a water distribution network into several district-metered areas (DMAs) yields is of great benefits for the management of water distribution systems, including enhanced leak control, pollution monitoring, and pressure optimization. However, achieving an effective partitioning that simultaneously incorporates node characteristics and network connections is challenging. Previous studies have tackled water network partitioning by focusing on either node features or pipe connections individually. In response to the limitations of current approaches for water distribution network partitioning, this study introduces an innovative unsupervised clustering approach for node partitioning based on graph convolutional network (GCN) techniques. Initially, the network nodes are clustered using the Cross-Attention Fusion Based Enhanced Graph Convolutional Network (CaEGCN) deep clustering algorithm, considering both the network’s topological layout and hydraulic properties. The resulting clusters categorize the network nodes into distinct regions. We evaluate our approach and other methods through a real world water distribution network using five different metrics commonly used to evaluate applied in assessing water supply network zoning issues. By comparing and analyzing the performance of experiments with other schemes, the experimental results demonstrate that the CaEGCN deep clustering method featured a remarkable performance. The proposed method showcases its potential for practical application by offering a dependable and easily interpretable water network partitioning solution.

ScEGG: an exogenous gene-guided clustering method for single-cell transcriptomic data.

In recent years, there has been significant advancement in the field of single-cell data analysis, particularly in the development of clustering methods. Despite these advancements, most algorithms continue to focus primarily on analyzing the provided single-cell matrix data. However, within medical contexts, single-cell data often encompasses a wealth of exogenous information, such as gene networks. Overlooking this aspect could result in information loss and produce clustering outcomes lacking significant clinical relevance. To address this limitation, we introduce an innovative deep clustering method for single-cell data that leverages exogenous gene information to generate discriminative cell representations. Specifically, an attention-enhanced graph autoencoder has been developed to efficiently capture topological signal patterns among cells. Concurrently, a random walk on an exogenous protein-protein interaction network enabled the acquisition of the gene's embeddings. Ultimately, the clustering process entailed integrating and reconstructing gene-cell cooperative embeddings, which yielded a discriminative representation. Extensive experiments have demonstrated the effectiveness of the proposed method. This research provides enhanced insights into the characteristics of cells, thus laying the foundation for the early diagnosis and treatment of diseases. The datasets and code can be publicly accessed in the repository at https://github.com/DayuHuu/scEGG.

NsDCC: dual-level contrastive clustering with nonuniform sampling for scRNA-seq data analysis.

Dimensionality reduction and clustering are crucial tasks in single-cell RNA sequencing (scRNA-seq) data analysis, treated independently in the current process, hindering their mutual benefits. The latest methods jointly optimize these tasks through deep clustering. However, contrastive learning, with powerful representation capability, can bridge the gap that common deep clustering methods face, which requires pre-defined cluster centers. Therefore, a dual-level contrastive clustering method with nonuniform sampling (nsDCC) is proposed for scRNA-seq data analysis. Dual-level contrastive clustering, which combines instance-level contrast and cluster-level contrast, jointly optimizes dimensionality reduction and clustering. Multi-positive contrastive learning and unit matrix constraint are introduced in instance- and cluster-level contrast, respectively. Furthermore, the attention mechanism is introduced to capture inter-cellular information, which is beneficial for clustering. The nsDCC focuses on important samples at category boundaries and in minority categories by the proposed nearest boundary sparsest density weight assignment algorithm, making it capable of capturing comprehensive characteristics against imbalanced datasets. Experimental results show that nsDCC outperforms the six other state-of-the-art methods on both real and simulated scRNA-seq data, validating its performance on dimensionality reduction and clustering of scRNA-seq data, especially for imbalanced data. Simulation experiments demonstrate that nsDCC is insensitive to "dropout events" in scRNA-seq. Finally, cluster differential expressed gene analysis confirms the meaningfulness of results from nsDCC. In summary, nsDCC is a new way of analyzing and understanding scRNA-seq data.

Automatic deep sparse clustering with a dynamic population-based evolutionary algorithm using reinforcement learning and transfer learning

Clustering data effectively remains a significant challenge in machine learning, particularly when the optimal number of clusters is unknown. Traditional deep clustering methods often struggle with balancing local and global search, leading to premature convergence and inefficiency. To address these issues, we introduce ADSC-DPE-RT (Automatic Deep Sparse Clustering with a Dynamic Population-based Evolutionary Algorithm using Reinforcement Learning and Transfer Learning), a novel deep clustering approach. ADSC-DPE-RT builds on Multi-Trial Vector-based Differential Evolution (MTDE), an algorithm that integrates sparse auto-encoding and manifold learning to enable automatic clustering without prior knowledge of cluster count. However, MTDE's fixed population size can lead to either prolonged computation or premature convergence. Our approach introduces a dynamic population generation technique guided by Reinforcement Learning (RL) and Markov Decision Process (MDP) principles. This allows for flexible adjustment of population size, preventing premature convergence and reducing computation time. Additionally, we incorporate Generative Adversarial Networks (GANs) to facilitate dynamic knowledge transfer between MTDE strategies, enhancing diversity and accelerating convergence towards the global optimum. This is the first work to address the dynamic population issue in deep clustering through RL, combined with Transfer Learning to optimize evolutionary algorithms. Our results demonstrate significant improvements in clustering performance, positioning ADSC-DPE-RT as a competitive alternative to state-of-the-art deep clustering methods.

Deep time-series clustering via latent representation alignment

In practice, obtaining sufficient label information from a dataset is challenging. Consequently, various clustering methods have been studied to homogeneously group data without label information. Recently, deep clustering approaches that utilize deep neural networks have garnered considerable attention. However, time series data possess unique characteristics, including temporal relationships between observations in a sequence, which can decrease the performance of existing deep clustering methods when applied to time series. Despite this, few studies on deep clustering have addressed the characteristics of time series. Thus, we propose a novel approach for deep time-series clustering using topological information, enabling the capture of underlying temporal patterns to generate cluster-oriented representations. We address the topological information of a time series by introducing a novel loss function based on the eigendecomposition of representations in latent space. Through experiments on various time-series datasets, we demonstrate the efficacy of the proposed method in achieving superior clustering performance compared to state-of-the-art deep clustering methods. To the best of our knowledge, this is the first approach that utilizes topological information for deep time-series clustering.

Developing deep learning models for predicting urban bike-sharing usage patterns

Urban traffic systems are facing significant challenges due to the ever-growing number of vehicles on the road, leading to increased congestion and suboptimal traffic flow. Traditional research focusing on individual traffic flows is often insufficient to meet the complex demands of modern urban transportation. While studying integrated shared single-vehicle flows offers a potential solution to mitigate these issues, the unique characteristics of shared bikes present substantial obstacles to accurate traffic flow research. These obstacles include the high liquidity, sparsity, and variability of shared bikes, the vagueness of travel characteristics, the lack of correlation between travel groups, and the unpredictability of travel patterns. The study endeavors to confront the challenges above by proposing an innovative model that correlates multiuser interactions and elucidates behavioral dynamics. This model utilizes a deep clustering method to analyze the evolution of superlarge-scale shared bike systems in Beijing. It uncovers the complex mechanisms governing user behavior and employs a neural network algorithm to predict shared bike users’ travel patterns effectively. By focusing on the theoretical and algorithmic aspects of behavioral dynamics for large-scale shared single-vehicle flows, this study offers a unique contribution to the field, with significant implications for multi-traffic flow management and urban planning in scenarios with extensive multi-traffic flows.

ScLEGA: an attention-based deep clustering method with a tendency for low expression of genes on single-cell RNA-seq data.

Single-cell RNA sequencing (scRNA-seq) enables the exploration of biological heterogeneity among different cell types within tissues at a resolution. Inferring cell types within tissues is foundational for downstream research. Most existing methods for cell type inference based on scRNA-seq data primarily utilize highly variable genes (HVGs) with higher expression levels as clustering features, overlooking the contribution of HVGs with lower expression levels. To address this, we have designed a novel cell type inference method for scRNA-seq data, termed scLEGA. scLEGA employs a novel zero-inflated negative binomial (ZINB) loss function that fully considers the contribution of genes with lower expression levels and combines two distinct scRNA-seq clustering strategies through a multi-head attention mechanism. It utilizes a low-expression optimized denoising autoencoder, based on the novel ZINB model, to extract low-dimensional features and handle dropout events, and a GCN-based graph autoencoder (GAE) that leverages neighbor information to guide dimensionality reduction. The iterative fusion of denoising and topological embedding in scLEGA facilitates the acquisition of cluster-friendly cell representations in the hidden embedding, where similar cells are brought closer together. Compared to 12 state-of-the-art cell type inference methods on 15 scRNA-seq datasets, scLEGA demonstrates superior performance in clustering accuracy, scalability, and stability. Our scLEGA model codes are freely available at https://github.com/Masonze/scLEGA-main.

A particle swarm optimization-based deep clustering algorithm for power load curve analysis

To address the inflexibility of the convolutional autoencoder (CAE) in adjusting the network structure and the difficulty of accurately delineating complex class boundaries in power load data, a particle swarm optimization deep clustering method (DC-PSO) is proposed. First, a particle swarm optimization algorithm for automatically searching the optimal network architecture and hyperparameters of CAE (AHPSO) is proposed to obtain better reconstruction performance. Then, an end-to-end deep clustering model based on a reliable sample selection strategy is designed for the deep clustering algorithm to accurately delineate the category boundaries and further improve the clustering effect. The experimental results show that the DC-PSO algorithm exhibits high clustering accuracy and higher performance for the power load profile clustering.

Clustering single-cell RNA sequencing data via iterative smoothing and self-supervised discriminative embedding.

Single-cell transcriptome sequencing (scRNA-seq) is a high-throughput technique used to study gene expression at the single-cell level. Clustering analysis is a commonly used method in scRNA-seq data analysis, helping researchers identify cell types and uncover interactions between cells. However, the choice of a robust similarity metric in the clustering procedure is still an open challenge due to the complex underlying structures of the data and the inherent noise in data acquisition. Here, we propose a deep clustering method for scRNA-seq data called scRISE (scRNA-seq Iterative Smoothing and self-supervised discriminative Embedding model) to resolve this challenge. The model consists of two main modules: an iterative smoothing module based on graph autoencoders designed to denoise the data and refine the pairwise similarity in turn to gradually incorporate cell structural features and enrich the data information; and a self-supervised discriminative embedding module with adaptive similarity threshold for partitioning samples into correct clusters. Our approach has shown improved quality of data representation and clustering on seventeen scRNA-seq datasets against a number of state-of-the-art deep learning clustering methods. Furthermore, utilizing the scRISE method in biological analysis against the HNSCC dataset has unveiled 62 informative genes, highlighting their potential roles as therapeutic targets and biomarkers.

1DCAE-TSSAMC: Two-Stage Multi-Dimensional Spatial Features Based Multi-View Deep Clustering for Time Series Data

At present, as a research hotspot for time series data (TSD), the deep clustering analysis of TSD has huge research value and practical significance. However, there still exist the following three problems: (1) For deep clustering based on joint optimization, the inevitably mutual interference existing between deep feature representation learning progress and clustering progress leads to difficult model training especially in the initial stage, the possible feature space distortion, inaccurate and weak feature representation; (2) Existing deep clustering methods are difficult to intuitively define the similarity of time series and rely heavily on complex feature extraction networks and clustering algorithms. (3) Multidimensional time series have the characteristics of high dimensions, complex relationships between dimensions, and variable data forms, thus generating a huge feature space. It is difficult for existing methods to select discriminative features, resulting in generally low accuracy of methods. Accordingly, to address the above three problems, we proposed a novel general two-stage multi-dimensional spatial features based multi-view deep clustering method 1DCAE-TSSAMC (One-dimensional deep convolutional auto-encoder based two-stage stepwise amplification multi-clustering). We conducted verification and analysis based on real-world important multi-scenario, and compared with many other benchmarks ranging from the most classic approaches such as K-means and Hierarchical to the state-of-the-art approaches based on deep learning such as Deep Temporal Clustering (DTC) and Temporal Clustering Network (TCN). Experimental results show that the new method outperforms the other benchmarks, and provides more accurate, richer, and more reliable analysis results, more importantly, with significant improvement in accuracy and spatial linear separability.

Robust deep image clustering using convolutional autoencoder with separable discrete Krawtchouk and Hahn orthogonal moments

By cooperatively learning features and assigning clusters, deep clustering is superior to conventional clustering algorithms. Numerous deep clustering algorithms have been developed for a variety of application levels; however, the majority are still incapable of learning robust noise-resistant latent features, which limits the clustering performance. To address this open research challenge, we introduce, for the first time, a new approach called: Robust Deep Embedded Image Clustering algorithm with Separable Krawtchouk and Hahn Moments (RDEICSKHM). Our approach leverages the advantages of Krawtchouk and Hahn moments, such as local feature extraction, discrete orthogonality, and noise tolerance, to obtain a meaningful and robust image representation. Moreover, we employ LayerNormalization to further improve the latent space quality and facilitate the clustering process. We evaluate our approach on four image datasets: MNIST, MNIST-test, USPS, and Fashion-MNIST. We compare our method with several deep clustering methods based on two metrics: clustering accuracy (ACC) and normalized mutual information (NMI). The experimental results show that our method achieves superior or competitive performance on all datasets, demonstrating its effectiveness and robustness for deep image clustering.

Deep Adaptive Fuzzy Clustering for Evolutionary Unsupervised Representation Learning.

Cluster assignment of large and complex datasets is a crucial but challenging task in pattern recognition and computer vision. In this study, we explore the possibility of employing fuzzy clustering in a deep neural network framework. Thus, we present a novel evolutionary unsupervised learning representation model with iterative optimization. It implements the deep adaptive fuzzy clustering (DAFC) strategy that learns a convolutional neural network classifier from given only unlabeled data samples. DAFC consists of a deep feature quality-verifying model and a fuzzy clustering model, where deep feature representation learning loss function and embedded fuzzy clustering with the weighted adaptive entropy is implemented. We joint fuzzy clustering to the deep reconstruction model, in which fuzzy membership is utilized to represent a clear structure of deep cluster assignments and jointly optimize for the deep representation learning and clustering. Also, the joint model evaluates current clustering performance by inspecting whether the resampled data from estimated bottleneck space have consistent clustering properties to improve the deep clustering model progressively. Experiments on various datasets show that the proposed method obtains a substantially better performance for both reconstruction and clustering quality compared to the other state-of-the-art deep clustering methods, as demonstrated with the in-depth analysis in the extensive experiments.

An overview on deep clustering

In recent years, with the great success of deep learning and especially deep unsupervised learning, many deep architectural clustering methods, collectively known as deep clustering, have emerged. Deep clustering shows the potential to outperform traditional methods, especially in handling complex high-dimensional data, taking full advantage of deep learning. To achieve a comprehensive overview of the field of deep clustering, this review systematically explores deep clustering methods and their various applications. First, the basic network architecture of deep clustering is described in detail, including the common network frameworks, and loss functions. Subsequently, deep clustering is divided into several categories based on the network architecture, and benchmark datasets and evaluation metrics in the field are introduced. Next, the real-world applications of deep clustering are explored in depth, providing successful cases in the fields of bioinformatics, medicine, anomaly detection, and image processing, highlighting the broad applicability of deep clustering in solving real-world challenges. Finally, the paper summarizes its contributions and explores potential directions for future research in deep clustering.

An End-to-end Deep Clustering Method with Consistency and Complementarity Attention Mechanism for Multisensor Fault Diagnosis

Deep clustering methods have found successful applications in single-sensor data fault diagnosis. However, most of these methods employ separate optimization strategies that overlook the interaction between feature learning and clustering. Moreover, conventional deep learning methods for fault diagnosis often disregard the consistent and complementary information inherent in the multisensor data, resulting in unsatisfactory multisensor fault diagnosis performance. In this study, we introduce a novel End-to-end Deep Clustering Method with Consistency and Complementarity Attention Mechanism, termed EDCM-CCAM, tailored for multisensor fault diagnosis. Firstly, multiple deep autoencoder networks are utilized to concurrently extract the deep representation features from various sensor inputs. Secondly, we introduce a Consistency and Complementarity Attention Mechanism (CCAM) to facilitate multisensor feature fusion, accompanied by the design of two distinct loss functions to fully exploit the consistent and complementary information within multisensor data. Finally, fault pattern recognition in multisensor data is accomplished through Kullback-Leibler (KL) divergence-based clustering, while a joint optimization strategy is employed to simultaneously optimize all components of the EDCM-CCAM. The efficacy of the proposed method is validated on a gearbox dataset, demonstrating superior performance in multisensor fault diagnosis compared to alternative methods.

Parallelly Adaptive Graph Convolutional Clustering Model.

Benefiting from exploiting the data topological structure, graph convolutional network (GCN) has made considerable improvements in processing clustering tasks. The performance of GCN significantly relies on the quality of the pretrained graph, while the graph structures are often corrupted by noise or outliers. To overcome this problem, we replace the pre-trained and fixed graph in GCN by the adaptive graph learned from the data. In this article, we propose a novel end-to-end parallelly adaptive graph convolutional clustering (AGCC) model with two pathway networks. In the first pathway, an adaptive graph convolutional (AGC) module alternatively updates the graph structure and the data representation layer by layer. The updated graph can better reflect the data relationship than the fixed graph. In the second pathway, the auto-encoder (AE) module aims to extract the latent data features. To effectively connect the AGC and AE modules, we creatively propose an attention-mechanism-based fusion (AMF) module to weight and fuse the data representations of the two modules, and transfer them to the AGC module. This simultaneously avoids the over-smoothing problem of GCN. Experimental results on six public datasets show that the effectiveness of the proposed AGCC compared with multiple state-of-the-art deep clustering methods. The code is available at https://github.com/HeXiax/AGCC.

Recognition and optimisation method of impact deformation patterns based on point cloud and deep clustering: Applied to thin-walled tubes

The recognition and clustering of deformation modes are key to constructing impact deformation constraints for thin-walled structures. This paper transforms the clustering and recognition problem of structural impact deformation modes into a problem of clustering and recognition of point cloud sequences based on pseudo-labels. The effectiveness of the method is assessed, and the experimental results show that the accuracy of the proposed method can reach up to 92.17 % when using a pre-training deep neural network feature extractor, which is not only close to the 98.50 % accuracy of supervised learning classification models but also has a 16.84 % improvement in accuracy compared to the deep clustering method based on K-Means. Under different clustering conditions, the proposed method can effectively classify and recognise samples with similar deformation modes and has the ability to summarise and induce new deformation modes when the number of clusters exceeds the number of manual labels. Furthermore, this paper presents a multi-objective optimisation method for structural crashworthiness under impact deformation constraints based on the NSGA-II algorithm. This method constructs impact deformation constraints using surrogate models and deformation clustering and recognition models. The experimental results show that the proposed method can effectively constrain the generation of the population. It is found that there are a large number of Pareto solutions that do not belong to the expected impact deformation mode under the condition of no deformation mode constraint. In contrast, almost all the obtained Pareto solutions conform to the expected impact deformation mode under the condition of deformation mode constraint. In summary, under the condition of impact deformation constraint, the obtained Pareto solutions can satisfy the crashworthiness requirements while conforming to the expected impact deformation mode.

Semi-supervised deep learning for molecular clump verification

Context. A reliable molecular clump detection algorithm is essential for studying these clumps. Existing detection algorithms for molecular clumps still require that detected candidates be verified manually, which is impractical for large-scale data. Semi-supervised learning methods, especially those based on deep features, have the potential to accomplish the task of molecular clump verification thanks to the powerful feature extraction capability of deep networks. Aims. Our main objective is to develop an automated method for the verification of molecular clump candidates. This method utilises a 3D convolutional neural network (3D CNN) to extract features of molecular clumps and employs semi-supervised learning to train the model, with the aim being to improve its generalisation ability and data utilisation. It addresses the issue of insufficient labelled samples in traditional supervised learning and enables the model to better adapt to new, unlabelled samples, achieving high accuracy in the verification of molecular clumps. Methods. We propose SS-3D-Clump, a semi-supervised deep clustering method that jointly learns the parameters of a 3D CNN and the cluster assignments of the generated features for automatic verification of molecular clumps. SS-3D-Clump iteratively classifies the features with the Constrained-KMeans and uses these class labels as supervision to update the weights of the entire network. Results. We used CO data from the Milky Way Imaging Scroll Painting project covering 350 square degrees in the Milky Way’s first, second, and third quadrants. The ClumpFind algorithm was applied to extract molecular clump candidates in these regions, which were subsequently verified using SS-3D-Clump. The SS-3D-Clump model, trained on a dataset comprising three different density regions, achieved an accuracy of 0.933, a recall rate of 0.955, a precision rate of 0.945, and an F1 score of 0.950 on the corresponding test dataset. These results closely align with those obtained through manual verification. Conclusions. Our experiments demonstrate that the SS-3D-Clump model achieves high accuracy in the automated verification of molecular clumps. It effectively captures the essential features of the molecular clumps and overcomes the challenge of limited labelled samples in supervised learning by using unlabelled samples through semi-supervised learning. This enhancement significantly improves the generalisation capability of the SS-3D-Clump model, allowing it to adapt effectively to new and unlabelled samples. Consequently, SS-3D-Clump can be integrated with any detection algorithm to create a comprehensive framework for the automated detection and verification of molecular clumps.

Simcryocluster: a semantic similarity clustering method of cryo-EM images by adopting contrastive learning

BackgroundCryo-electron microscopy (Cryo-EM) plays an increasingly important role in the determination of the three-dimensional (3D) structure of macromolecules. In order to achieve 3D reconstruction results close to atomic resolution, 2D single-particle image classification is not only conducive to single-particle selection, but also a key step that affects 3D reconstruction. The main task is to cluster and align 2D single-grain images into non-heterogeneous groups to obtain sharper single-grain images by averaging calculations. The main difficulties are that the cryo-EM single-particle image has a low signal-to-noise ratio (SNR), cannot manually label the data, and the projection direction is random and the distribution is unknown. Therefore, in the low SNR scenario, how to obtain the characteristic information of the effective particles, improve the clustering accuracy, and thus improve the reconstruction accuracy, is a key problem in the 2D image analysis of single particles of cryo-EM.ResultsAiming at the above problems, we propose a learnable deep clustering method and a fast alignment weighted averaging method based on frequency domain space to effectively improve the class averaging results and improve the reconstruction accuracy. In particular, it is very prominent in the feature extraction and dimensionality reduction module. Compared with the classification method based on Bayesian and great likelihood, a large amount of single particle data is required to estimate the relative angle orientation of macromolecular single particles in the 3D structure, and we propose that the clustering method shows good results.ConclusionsSimcryoCluster can use the contrastive learning method to perform well in the unlabeled high-noise cryo-EM single particle image classification task, making it an important tool for cryo-EM protein structure determination

Self-Supervised Deep Clustering Method for Detecting Abnormal Data of Wastewater Treatment Process

In wastewater treatment process (WWTP), abnormal data seriously reduce data quality rendering the application techniques impractical. The implementation of abnormal data detection is challenging due to the nonlinear nature of WWTP. Typically, constructing an accurate anomaly detector requires large amounts of labeled data, which is difficult in practice. Thus, a self-supervised memory enhanced deep clustering method (SMEL) is proposed to detect abnormal data without using any labels. First, a self-supervised deep clustering network, combining stacked autoencoders and the clustering algorithm, leverages unlabeled data to extract nonlinear features and capture the normal pattern. Second, an adaptive weight objective function, jointly optimizing the reconstruct error and clustering error, is designed to obtain a robust clustering structure. Third, double memory enhanced modules, consisting of a centroid memory and a score memory, are presented to enhance training stability and detection accuracy. Finally, experiments on three WWTP datasets show that SMEL achieves the highest detection accuracy.