Improved Fuzzy C-Means Research Articles

Enhanced gastric cancer classification and quantification interpretable framework using digital histopathology images

Recent developments have highlighted the critical role that computer-aided diagnosis (CAD) systems play in analyzing whole-slide digital histopathology images for detecting gastric cancer (GC). We present a novel framework for gastric histology classification and segmentation (GHCS) that offers modest yet meaningful improvements over existing CAD models for GC classification and segmentation. Our methodology achieves marginal improvements over conventional deep learning (DL) and machine learning (ML) models by adaptively focusing on pertinent characteristics of images. This contributes significantly to our study, highlighting that the proposed model, which performs well on normalized images, is robust in certain respects, particularly in handling variability and generalizing to different datasets. We anticipate that this robustness will lead to better results across various datasets. An expectation-maximizing Naïve Bayes classifier that uses an updated Gaussian Mixture Model is at the heart of the suggested GHCS framework. The effectiveness of our classifier is demonstrated by experimental validation on two publicly available datasets, which produced exceptional classification accuracies of 98.87% and 97.28% on validation sets and 98.47% and 97.31% on test sets. Our framework shows a slight but consistent improvement over previously existing techniques in gastric histopathology image classification tasks, as demonstrated by comparative analysis. This may be attributed to its ability to capture critical features of gastric histopathology images better. Furthermore, using an improved Fuzzy c-means method, our study produces good results in GC histopathology picture segmentation, outperforming state-of-the-art segmentation models with a Dice coefficient of 65.21% and a Jaccard index of 60.24%. The model’s interpretability is complemented by Grad-CAM visualizations, which help understand the decision-making process and increase the model’s trustworthiness for end-users, especially clinicians.

Adaptive load balancing in distributed cloud environment: Hybrid Kookaburra-Osprey optimization algorithm

A distributed cloud environment is characterized by the dispersion of computing resources, services, and applications across multiple locations or data centres. This distribution enhances scalability, redundancy, and resource utilization efficiency. To optimize performance and prevent any single node from becoming a bottleneck, it is imperative to implement effective load-balancing strategies, particularly as user demands vary and certain nodes experience increased processing requirements. This research introduces an Adaptive Load Balancing (ALB) approach aimed at maximizing the efficiency and reliability of distributed cloud environments. The approach employs a three-step process: Chunk Creation, Task Allocation, and Load Balancing. In the Chunk Creation step, a novel Improved Fuzzy C-means clustering (IFCMC) clustering method categorizes similar tasks into clusters for assignment to Physical Machines (PMs). Subsequently, a hybrid optimization algorithm called the Kookaburra-Osprey Updated Optimization Algorithm (KOU), incorporating the Kookaburra Optimization Algorithm (KOA) and Osprey Optimization Algorithm (OOA), allocates tasks assigned to PMs to Virtual Machines (VMs) in the Task Allocation step, considering various constraints. The Load Balancing step ensures even distribution of tasks among VMs, considering migration cost and efficiency. This systematic approach, by efficiently distributing tasks across VMs within the distributed cloud environment, contributes to enhanced efficiency and scalability. Further, the contribution of the ALB approach in enhancing the efficiency and scalability of distributed cloud environments is evaluated through analyses. The KBA is 1189.279, BES is 629.240, ACO is 1017.889, Osprey is 1147.300, SMO is 1215.148, APDPSO is 1191.014, and DGWO is 1095.405, respectively. The resource utilization attained by the KOU method is 1224.433 at task 1000.

Logging Identification Method for Reservoir Facies in Fractured-Vuggy Dolomite Reservoirs Based on AI: A Case Study of Ediacaran Dengying Formation, Sichuan Basin, China

As an important target for deep to ultra-deep carbonate oil and gas exploration, Fractured-Vuggy dolomite reservoirs have strong heterogeneity. Accurate characterization of reservoir facies is crucial for their exploration and exploitation. Three methods, including the unsupervised intelligent clustering method of improved Fuzzy C-means clustering Algorithm Based on Density Sensitive Distance and Fuzzy Partrition (FCM-DSDFP), the fusion method of Principal Components Analysis (PCA) dimensionality reduction and noise reduction, and the principle of clustering feature analysis are applied to identify reservoir facies based on logging data. Based on the PCA method, the logging response characteristics of the reservoir facies are excavated, and the fusion characterization data of dimensionality reduction and noise reduction are extracted. The FCM-DSDFP unsupervised intelligent clustering method, a model that approximates the subsurface conditions is established, and the reliability of the model is tested according to the elbow rule and silhouette coefficient. Combining drilling core observation, Fractured-Vuggy type, partially cemented Fractured-Vuggy type, Pore-Vuggy type, Pore Type I, Pore Type II, Tight Type I, and Tight Type II are divided in the Dengying Formation 4th Member. Fractured-Vuggy type, partially cemented Fractured-Vuggy type, Pore-Vuggy Type I, Pore-Vuggy Type II, Pore Type I, Pore Type II, and Tight Type are divided in the Dengying Formation 2nd Member, respectively. Two methods were applied to verify the reservoir facies identification results based on intelligent algorithms. The first method is to compare the identification results with the reservoir facies types identified by core observations (Well PT103 and PS13). The second method is to verify the recognition results of intelligent algorithms by utilizing the relationship between reservoir facies types and bitumen. The test results show that the accuracy of the reservoir level identification is higher than 0.9, and the applicability is better than the commonly used algorithms such as FCM and K-means.

An Intelligent Moving Object Segmentation Using Hybrid IFCM-CSS Clustering Model

In this study, a novel hybrid deep clustering approach is proposed for the effective moving object segmentation. Initially, the data is collected, and the keyframe selection is performed using the threshold-based Kennard–Stone method. Then, the preprocessing step involves noise filtering using bilateral wavelet thresholding and binary color conversion. The blob detection is performed using normalized Laplacian of Gaussian. Finally, the segmentation of moving objects is performed using a hybrid clustering approach called improved fuzzy C-mean (IFCM) clustering with chaotic salp swarm (CSS) optimization algorithm (Hybrid IFCM-CSS). The overall evaluation is done in MATLAB. The performance of the hybrid IFCM-CSS is compared to other approaches based on some measures. The proposed Hybrid IFCM-CSS achieves the highest precision of 0.971, using the SBM-RGBD dataset.

Advanced Integration of Machine Learning Techniques for Accurate Segmentation and Detection of Alzheimer’s Disease

Alzheimer’s disease is a common type of neurodegenerative condition characterized by progressive neural deterioration. The anatomical changes associated with individuals affected by Alzheimer’s disease include the loss of tissue in various areas of the brain. Magnetic Resonance Imaging (MRI) is commonly used as a noninvasive tool to assess the neural structure of the brain for diagnosing Alzheimer’s disease. In this study, an integrated Improved Fuzzy C-means method with improved watershed segmentation was employed to segment the brain tissue components affected by this disease. These segmented features were fed into a hybrid technique for classification. Specifically, a hybrid Convolutional Neural Network–Long Short-Term Memory classifier with 14 layers was developed in this study. The evaluation results revealed that the proposed method achieved an accuracy of 98.13% in classifying segmented brain images according to different disease severities.

EM-IFCM: Fuzzy c-means clustering algorithm based on edge modification for imbalanced data

The improved fuzzy c-means (IFCM) algorithm is an effective technique for handling the “uniform effect” in imbalanced data clustering; it adjusts the weight of each class based on the fuzzy size between clusters. However, the IFCM algorithm produces a “siphon effect” as the imbalance rate increases. It misclassifies the samples in small classes into large ones. Our analysis shows that this effect occurs because all samples have the same weight value of the same classes, the membership values are polarized, resulting in the model failing to converge to the correct interval. Thus, we propose an imbalanced fuzzy c-means clustering based on edge modification (EM-IFCM) algorithm to alleviate the “siphon effect” of the IFCM algorithm. It exhibits stronger inter-class separability by dynamically adjusting the weight of the samples to enhance the influence of edge samples on the model. In addition, we analyze the effectiveness and complexity of the algorithm and proved its convergence. Finally, we conduct extensive experiments on synthesis, machine-learning, and image-segmentation datasets and compare the results with those of six algorithms. The experimental results show that EM-IFCM has higher accuracy and exhibits an imbalance rate that is at least 1.94 times higher than that of the other algorithms.

Module division method of complex products for responding to user’s requirements

With the gradual diversification of personalized usage scenarios, user requirements play a direct role in product design decisions. Due to the problem of fuzzy demand caused by user cognitive bias, traditional design methods usually focus on realizing product functions and cannot effectively match user requirements. Therefore, this paper proposes a complex product module division method for user requirements. The method constitutes of three tasks, requirement analysis of module division, design mapping of module division and scheme implementation of module division. Firstly, based on the progressive architecture from initial requirements to precise requirements, the effective user requirements are obtained through similarity recommendation. Secondly, according to the four types of knowledge of function, geometry, physics and design, the design structure matrix is constructed to complete the Requirement-Function-Structure mapping. The improved Fuzzy C-means Algorithm is used to solve the mapping model, and finally a module division scheme that meets the user requirements is obtained. Taking the chip removal machine as an example, the rationality and effectiveness of the method are verified. The results show that the product modules divided by this method can effectively meet the multiple user requirements.

Detection and location of microaneurysms in fundus images based on improved YOLOv4 with IFCM

AbstractDiabetic retinopathy (DR) is one of the leading causes of blindness for people suffering from diabetes. Microaneurysm (MA) is the initial symptom of DR. MA is a lesion based disease which starts as small red spots on the retina and increases in size as the DR progresses which finally leads to blindness. So eliminating the lesion can effectively prevent DR at an early stage. However, due to complex retinal structure, different brightness and contrast of fundus images with effects of factors such as different patients, environment changes, and difference in acquisition equipment, it is difficult for existing detection algorithms to achieve accurate results of MA detection and location. Therefore, the detection algorithm of improved YOLOv4 (YOLOv4‐Pro) was proposed. First, an improved Fuzzy C‐Means (IFCM) clustering algorithm was proposed to optimize anchor parameters of target samples to improve matching results between anchors and feature graphs. In order to control noise and improve efficiency, a median filtering method was employed to update the criterion function of the original FCM algorithm, and a K‐means algorithm was employed to initialize clustering. Second, a SENet attention module was added in the backbone of YOLOv4 to enhance key information and suppress background, improving the confidence of MA effectively. Finally, the spatial pyramid pooling (SPP) module was added to the neck to enhance the acceptance domain of the output characteristics of the backbone network, and profits separating of important context information. The improved YOLOv4 with IFCM was verified on the Kaggle DR dataset and compared with other methods. Experimental results show that optimizing the prior frame with the IFCM algorithm can make it suitable to frame the Kaggle DR dataset, which improves the detection accuracy of the network by nearly 5%, and provides a nice performance on detection and location of MA in fundus images. This would help ophthalmologists finding the exact location of MA on retina, thereby simplifying the process and eliminating any manual intervention.

An improved fuzzy C-means clustering algorithm using Euclidean distance function

The fuzzy c-mean (FCM) clustering algorithm is a typical algorithm using Euclidean distance for data clustering and it is also one of the most popular fuzzy clustering algorithms. However, FCM does not perform well in noisy environments due to its possible constraints. To improve the clustering accuracy of item varieties, an improved fuzzy c-mean (IFCM) clustering algorithm is proposed in this paper. IFCM uses the Euclidean distance function as a new distance measure which can give small weights to noisy data and large weights to compact data. FCM, possibilistic C-means (PCM) clustering, possibilistic fuzzy C-means (PFCM) clustering and IFCM are run to compare their clustering effects on several data samples. The clustering accuracies of IFCM in five datasets IRIS, IRIS3D, IRIS2D, Wine, Meat and Apple achieve 92.7%, 92.0%, 90.7%, 81.5%, 94.2% and 88.0% respectively, which are the highest among the four algorithms. The final simulation results show that IFCM has better robustness, higher clustering accuracy and better clustering centers, and it can successfully cluster item varieties.

Early Detection and Classification of Heart Diseases by Employing IFCMML and 2L-C Model with I-GA Machine Learning Methods

Objectives: To identify a new method for early detection and classification of Heart Diseases (HD); to improve the accuracy of the results by employing I-FCMML (Improved Fuzzy C-Means Machine Learning) and 2L-C (Two-Level Classifier) models along with the I-GA (Improved Genetic Algorithm) methods. Methods: The I-FCMML algorithm is utilized for feature selection and extraction. Machine learning techniques such as Ensembled Random Forest method (ERFM) and Robust Gradient Boost Method (RGBT) are employed to predict the likelihood of HD and 2L-C, I-GA is used to classify and detect the HD at a premature stage based on features like age, gender, blood pressure, etc. IFCMML with 2L-C & I-GA extracts all the features from the dataset (Cleveland HD dataset from the Cleveland Clinic Foundation, Ohio, USA) which includes 303 observations, 14 features and selects the suitable function to perform disease classification and detection with high accuracy. To evaluate the performance of the proposed method, MATLAB is used for implementation. The results are compared with existing algorithms such as 3P-ANN, ANN-FAHP, ADWFS, EDSS, and FE-PCA. Findings: Early HD detection and classification is achieved with 96.02% accuracy, 95.80% sensitivity, 94.76% specificity, 95% precision, 94% recall, 0.90 True Positive, 0.87 True Negative, and 94.13% F-Score to detect and classify the HD in a robust manner, which is comparatively high than the existing methods. Novelty: According to the findings of the comprehensive study, the proposed new method I-FCMML with 2L-C & I-GA has the potential to provide accurate and competent detection and classification of HD at an early stage, which could help for timely treatment and management of HD patients, and it also outperforms the existing methods such as 3P-ANN, ANNFAHP, ADWFS, EDSS, and FE-PCA. Keywords: Heart Disease Detection; Classification Algorithm; Genetic Algorithm; Fuzzy C-Means ML; Image Processing

Prediction method of wind farm power generation capacity based on feature clustering and correlation analysis

• A wind power forecasting method based on deep sparse autoencoder is proposed. • An improved fuzzy C-means clustering algorithm is proposed. • Combining IFCM with PCC optimizes the input data for model training. • A DSAE model is constructed to effectively improve the prediction accuracy. Accurate wind power forecasting can help power systems achieve economical operation and dispatch management. This paper proposes a short-term wind power forecasting method based on feature clustering and correlation analysis, improving forecasting accuracy through data feature clustering, variable correlation analysis, and building forecasting models. More specifically, the improved fuzzy C-means (IFCM) algorithm is used to cluster the wind power dataset; Pearson correlation coefficient (PCC) is used to explore the correlation between meteorology and wind power variables; Based on the deep sparse auto-encoder (DSAE) model, the predicted value of short-term wind power is obtained. Combined with the actual wind power data, the effectiveness and superiority of the proposed method are verified by comparison.

A novel Moore-Penrose pseudo-inverse weight-based Deep Convolution Neural Network for bacterial leaf blight disease detection system in rice plant

The vital food crop in the agriculture field is rice, but rice growing is impacted by numerous maladies. The Bacterial Leaf Blight (BLB) disease influences rice standards. Prevailing research practices have less accuracy and could not surmount the noise of the images. This paper proposed Moore-Penrose pseudo-inverse Weight-related Deep Convolutional Neural Network (MPW-DCNN) overcomes such complications for the BLB disease identification system. The proposed method encompasses six phases. First, the Image Acquisition (IA) procedure is accomplished, next the pre-processing is carried out, in which the noise of the input rice leaf image is mitigated using the Hybrid Gaussian-Weiner (HGW) filter, and the image pixel is normalized by utilizing min-max normalization. Then, the segmentation of the input image is done by Improved Fuzzy C-Means (IFCM). Next, the features, such as entropy, energy, correlation, contrast, homogeneity, colour histogram, and Scale-Invariant Feature Transforms (SIFT) are extracted from the segmented image. Then, the efficient features are chosen by applying the Exhaustiveness and Brownian Motion-related Elephant Herding Optimization (EBM-EHO) algorithm. Then, these selected features are fed to the MPW-DCNN classifier, which categorizes the image as 'BLB malady' or ‘normal’ or else 'chances being influenced by other maladies'. Lastly, the proposed MPW-DCNN performance is equated with the prevailing classification algorithm, and better accuracy is proffered by the proposed work than the prevalent algorithms. The accuracy of the implemented approach is 2.36, 3.08, 4.62, 5.13, and 6.15% improved than the existing methods, such as Support Vector Machine (SVM) + Deep features, AlexNet, Deep Convolutional Neural Network (DCNN), Convolutional Neural Network (CNN), and Artificial Neural Network (ANN), respectively.

Data Processing Method of Mine Wind Speed Monitoring Based on an Improved Fuzzy C-Means Clustering Algorithm

Analyzing and processing mine wind speed monitoring data is the key to realizing intelligent ventilation and real-time calculation of the ventilation network. According to the characteristics of the artificial regulation of a mine ventilation system, a local regression fuzzy C clustering algorithm is proposed in this paper, which combines local outlier processing with global air volume state analysis. Firstly, the algorithm uses the robust local weighted regression principle to analyze and preprocess the data locally, determines the risk degree of the abnormal data according to the identified times of outliers, determines the clustering number according to the clustering validity function, and analyzes the global air volume fluctuation according to the clustering results. The results show that most outliers are identified in data preprocessing. Still, the processing of dense outliers is weak, related to the window width setting and weighting multiple. The number of clusters can represent the fluctuation of the ventilation state and the pre-processed cluster centers are 4.4% lower than the original data because most of the outliers are higher than the average data. According to the law of air volume balance, the clustering results can pave the way for the global deduction of mine wind speed. There is an implicit relationship between data preprocessing and the clustering process, and when intensive outliers are not eliminated, they may be identified as separate clusters. The research of this paper points out the direction of mine wind speed data analysis, which can provide a theoretical basis for intelligent mine ventilation and real-time calculation of the ventilation network.

A Repair Method for Missing Traffic Data Based on FCM, Optimized by the Twice Grid Optimization and Sparrow Search Algorithms.

Complete traffic sensor data is a significant prerequisite for analyzing the changing rules of traffic flow and formulating traffic control strategies. Nevertheless, the missing traffic data are common in practice. In this study, an improved Fuzzy C-Means algorithm is proposed to repair missing traffic data, and three different repair modes are established according to the correlation of time, space, and attribute value of traffic flow. First, a Twice Grid Optimization (TGO) algorithm is proposed to provide a reliable initial clustering center for the FCM algorithm. Then the Sparrow Search Algorithm (SSA) is used to optimize the fuzzy weighting index m and classification number k of the FCM algorithm. Finally, an experimental test of the traffic sensor data in Shunyi District, Beijing, is employed to verify the effectiveness of the TGO-SSA-FCM. Experimental results showed that the improved algorithm had a better performance than some traditional algorithms, and different data repair modes should be selected under different miss rate conditions.

Image Enhancement Method based on an Improved Fuzzy C-Means Clustering

Image enhancement is an important method in the process of image processing. This paper proposes an image enhancement method base on an improved fuzzy c-means clustering. The method consists of the following steps: firstly, proposed a fuzzy c-means clustering with a cooperation center(FCM-co). Secondly, using the FCM-co, divide the image pixels into different clusters and marked membership values to those clusters. Thirdly, modify the membership values. Finally, calculate the new pixel gray levels. This enhancement method can overcome the disadvantage of overexposure and better retain image details. Through the experiment, the test results show that the proposed enhancement method could achieve better performance.

Optimal Excess Commuting Evaluation Based on Local Minimal Costs

Excess commuting refers to the value of unnecessary commuting or distance costs. Traditional commuting distance models adapt the most efficient scenario with people working in the nearest workplace geographically. Even though there have been some attempts to include constraints with commuter attributes and neighborhood features, problems arise with traditional geographical space and the subjectivity of these predefined characteristics. In this paper, we propose a method to calculate theoretical local minimal costs, which considers preferences that are inherently behavioral based on current work–home trips in the process of reassigning the work–home configuration. Our method is based on a feature space with a higher dimension and with the enlargement of attributes and relations of and between commuters and neighborhoods. Additionally, our solution is arrived at innovatively by improved Fuzzy C-Means clustering and linear programming. Unlike traditional clustering algorithms, our improved method adapts entropy information and selects the initial parameters based on the actual data rather than on prior knowledge. Using the real origin–destination matrix, theoretical minimal costs are calculated within each cluster, referred to as local minimal costs, and the average sum of local minimal costs is our theoretical minimal cost. The difference between the expected minimal cost and the actual cost is the excess commuting. Using our method, experimental results show that only 13% of the daily commuting distance in Wuhan could be avoided, and the theoretical distance is approximately 1.06 km shorter than the actual commuting distance.

Weighted Synthetic Minority Over-Sampling Technique (WSMOTE) Algorithm and Ensemble Classifier for Hepatocellular Carcinoma (HCC) In Liver Disease System

HCC (Hepato Cellular Carcinoma) is a generic liver cancer causing death in people suffering from LC (Liver Cirrhosis). Early prognosis of HCC is a significant factor to the line of LC treatment for clinicians. Though current treatments for HCC have been effective, patients have responded negatively or exhibited aggressive biological behaviours. Identification accuracy reduces when sub-optimal models of MLTs (Machine Learning Techniques) process multiple classes. MLT models base their predictions on their training phases where class imbalances in datasets have retained challenges in terms of unsatisfactory results. This research work attempts to overcome these issues with its proposed WSMOTE (Weighted Synthetic Minority Over-sampling Technique) algorithm which is targeted at dataset imbalances. Missing values are imputed using IFCM (Improved Fuzzy C Means) clustering for enhancing analysis accuracy. Imputed features are chosen selectively using IEFS (Intelligence Ensemble-Based Feature Selection). A heterogeneous ensemble classifier using Bootstrap aggregation is applied for combining predictions of multiple classifiers including KSVMs (Kernel Support Vector Machines) and FCNNs (Fuzzy Convolution Neural Networks) for accuracy in outputs. This work’s schema when tested on MATLAB (Matrix Laboratory) was found to classify better than most other methods in terms of precision, recall, F-measure and accuracy.

Improved Fuzzy C-Means for Infrared Small Target Detection

Infrared (IR) small target detection has been extensively studied due to its importance in IR search and tracking (IRST) systems. Existing methods have some limitations in suppressing cluttered high-contrast backgrounds, which may result in more false detections. In this letter, on the one hand, we propose an improved fuzzy C-mean (IFCM) clustering to accurately segment IR small targets and complex backgrounds. On the other hand, an IFCM-based descriptor fusing multiple features (IFCM-MF) is proposed to suppress complex backgrounds. First, a sliding window is designed to quickly extract the candidate pixels. Then, to better enhance the target and suppress the background, we construct an IFCM-based local window and calculate the IFCM-MF. Finally, IR small targets are detected by adaptive thresholding operation. The experimental results show that our method can better suppress cluttered high-contrast backgrounds and significantly improve the detection performance with high speed.

Crack Extraction for Polycrystalline Solar Panels

Crack extraction of solar panels has become a research focus in recent years. The cracks are small and hidden. In addition, there are particles of irregular shape and size on the surface of the polycrystalline solar panel, whose reflection position and direction are random. Therefore, there is a complex and uneven texture background on the solar panel image, which makes the crack extraction more difficult. In this paper, a crack extraction method combining image texture and morphological features is proposed. Firstly, the background texture and multi-scale details are suppressed by the linear filter and the Laplace pyramid decomposition method. Secondly, the edge can be extracted based on the modulus maximum method of the wavelet transform. Finally, cracks were extracted by using the improved Fuzzy C-means (FCM) clustering combining the morphological and texture features of the cracks. To make the extraction results more accurate and reasonable, an improved region growth algorithm is proposed to optimize the extraction results. All of the above research is closely centered on the accuracy and stability requirements of the solar cell crack detection, which is also the key point of this paper. The experimental results show that various improved or innovative algorithms proposed in this paper can accurately extract the position of cracks and obtain better extraction results. The detection results have good stability and can be faithful to the actual situation, which will promote the application of solar cells in more fields.

Day-ahead wind power forecasting based on the clustering of equivalent power curves

Wind power prediction (WPP) has developed in recent years into a way to solve the strong fluctuation problems that are caused by large-scale integration. Higher prediction accuracy is important to improve power grid security and economy. Wind turbine power curves, which describe the transformation between speed and power output, have been widely applied to WPP. In order to improve the accuracy of the prediction results and reduce the complexity of the model, this research proposes an improved Fuzzy C-means (FCM) Clustering Algorithm for day-ahead wind power prediction to resolve the difference in wind power output. By using the principle of minimum distance to select the relatively rough initial cluster centers of the samples, better clustering results can be obtained. The improved FCM method is used to classify turbines with similar power output characteristics into several categories, and a representative power curve is selected as the equivalent curve of the wind farm. And then capture the performance of the wind turbine. A day-ahead WPP model which utilizes numerical weather predictions (NWPs) as inputs for a subsequent equivalent power curve model is therefore established. The model proposed was validated using historical data taken from two different wind farms located in northeastern China.