Fuzzy C-means Method Research Articles

Perbandingan Metode K-Means Dan Metode Fuzzy C-Means (FCM) Pada Analisis Kinerja Pegawai PT. Cemara Khatulistiwa Persada Bontang

Discipline assessment of employee performance is one of the factors to improve the situation of the quality of human resources. Monitoring and assessment of employee discipline must be carried out continuously as some of the characteristics of management that have gone well as a benchmark for considering the targets that have been set. The K-Means method and the Fuzzy C-Means (FCM) method are non-hierarchical cluster methods. Both methods attempt to partition data into one or more clusters, so that data with the same characteristics are grouped into the same cluster or groups and data with different characteristics are grouped into other groups. This study discusses the comparison of the K-Means method and the Fuzzy C-Means (FCM) method in analyzing employee performance at PT. Cemara Khatulistiwa Persada Bontang, where groups of employees with high, medium, and low levels of employee performance will be determined based on the clustering results of the two methods and determine the best method. The grouping of data for the two methods was obtained from employee attendance data in 2020. Based on the results, it was found that clustering using the K-Means method in the first cluster (low performance level) had 23 employees, the second cluster (medium performance level) had 27 employees, and cluster the third (high performance level) there are 30 employees. Then based on the results of clustering using the FCM method in the first cluster (medium performance level) there are 26 employees, the second cluster (high performance level) there are 31 employees, and the third cluster (low performance level) there are 23 employees. Based on the results of the standard deviation ratio, it was obtained that the K-Means method with a value of 2.4944 was better than the FCM method with a value of 2.7323 in clustering employee performance at PT. Cemara Khatulistiwa Persada.

Application of Fuzzy C-Means and Weighted Scoring Methods for Mapping Blankspot Villages in Pemalang Regency

Covid-19 pandemic affects habits people around the world. The education sector in Indonesia is also undergoing policy changes, namely policy of transitioning face-to-face teaching and learning process to distance learning process (PJJ/online learning). Several studies have been conducted to examine the constraints PJJ process, resulting in finding that quality of internet network is majority obstacle in PJJ process. Conditions where there is no internet network in an area is commonly called a blankspot. In order to minimize the problem of blankspots, President and Ministry of Communication and Informatics of Indonesia realized the program "Indonesia is free signals to the corners of the country". This program involves all districts in Indonesia to conduct network quality surveys in the smallest areas of the village. Basically, network quality survey activities require relatively no small resources and costs. So as to conduct the efficiency of field survey activities, early detection of village blankspot status is required based on the characteristics blankspot village in general. While the commonly used method of grouping village based on village characteristics is the fuzzy c-means and weighted scoring method. These two methods were chosen because they have good cluster convergence rate and easily interpreted display results of the group by user in the form diagrams and scores. This study aims to prove that fuzzy c-means and weighted scoring method are good for grouping cases of blankspot villages according to previous studies with different cases. The result comparison goodness value of clustering, it is known that fuzzy c-means method more suitable for clustering characteristics blankspot village than the k-means method. Meanwhile, weighted scoring method cannot be said better method for village classification than the decision tree method.

In-situ manifestation of lapping mechanisms by rapid intelligent pattern recognition analysis (RIPRA) of acoustic emission via a point density fuzzy C-means (PD-FCM) method

Rapid pattern-recognition capabilities are very challenging tasks to fulfill while essential steps for the ultimate mission of achieving in-time security response, hazardous disaster prevention, structural failure prediction, and intelligent manufacturing applications. Mechanistic manifestation of a machining process through pattern recognition is even more daunting since one has to gain full understanding at the very fundamental level of the manufacturing processes, but it would be the most beneficial should it be accomplished. We designed and built an in-situ Acoustic Emission (AE) monitoring system, and successfully applied this newly developed AE system to manifest the lapping removal mechanisms of a novel fixed abrasive pad embedded with agglomerated diamond (AD) abrasives. To successfully extract valuable mechanistic information, the time-frequency spectrum of collected AE signals in a sapphire lapping process was decomposed by Wavelet Packet Transform (WPT) into three characteristically patterned bands, each representing rubbing, ploughing and cutting action in the lapping process correspondingly. We creatively applied Point Density Fuzzy C-Means (PD-FCM) algorithm for self-identifying clustering centers to classify contributions from rubbing, ploughing and cutting actions in the lapping process. By rapid intelligent pattern recognition analysis (RIPRA) of AE signals, the lapping removal mechanisms using fixed abrasive pads were clearly identified and quantitatively monitored in-situ throughout the whole lapping processes. Such a capable approach implicates a great potential in a wide range of ultra-precision manufacturing applications.

Botanical Leaf Disease Detection and Classification Using Convolutional Neural Network: A Hybrid Metaheuristic Enabled Approach

Botanical plants suffer from several types of diseases that must be identified early to improve the production of fruits and vegetables. Mango fruit is one of the most popular and desirable fruits worldwide due to its taste and richness in vitamins. However, plant diseases also affect these plants’ production and quality. This study proposes a convolutional neural network (CNN)-based metaheuristic approach for disease diagnosis and detection. The proposed approach involves preprocessing, image segmentation, feature extraction, and disease classification. First, the image of mango leaves is enhanced using histogram equalization and contrast enhancement. Then, a geometric mean-based neutrosophic with a fuzzy c-means method is used for segmentation. Next, the essential features are retrieved from the segmented images, including the Upgraded Local Binary Pattern (ULBP), color, and pixel features. Finally, these features are given into the disease detection phase, which is modeled using a Convolutional Neural Network (CNN) (deep learning model). Furthermore, to enhance the classification accuracy of CNN, the weights are fine-tuned using a new hybrid optimization model referred to as Cat Swarm Updated Black Widow Model (CSUBW). The new hybrid optimization model is developed by hybridizing the standard Cat Swarm Optimization Algorithm (CSO) and Black Widow Optimization Algorithm (BWO). Finally, a performance evaluation is undergone to validate the efficiency of the projected model.

Pengelompokan Jenis Penerimaan Pajak di Kota Makassar Menggunakan Fuzzy Clustering

This research discusses grouping types of tax receipts in the City of Makassar with the method of Fuzzy C-Means (FCM). The data used are secondary data from the tax office in the form of the amount of the receipts of income tax, value-added tax, and land tax building each district in the year 2018. The discussion starts from the theory of fuzzy clustering, simulation, and the characteristics of the simulation results. In this study, the obtained number of clusters the optimum, namely the 7 clusters that have the potential for tax revenue is relatively high, namely at cluster seven consists only of the District Tamalanrea, and the lowest in the cluster a sixth consisting of the Kecamatan Tallo, Ujung Tanah, and Sangkarrang. After hashing on a cluster that is formed can be used to see the level of the welfare of the community in each district.

IMPLEMENTASI ALGORITMA FUZZY C-MEANS DAN FUZZY POSSIBILISTICS C-MEANS UNTUK KLASTERISASI DATA TWEETS PADA AKUN TWITTER TOKOPEDIA

Social media has become the most popular media, which can be accessed by young to old age. Twitter became one of the effective media and the familiar one used by the public, thus making the company make Twitter one of the promotional tools, one of which is Tokopedia. The research aims to group tweets uploaded by @tokopedia Twitter accounts based on the type of tweets content that gets a lot of retweets and likes by followers of @tokopedia. Application of text mining to cluster tweets on the @tokopedia Twitter account using Fuzzy C-Means and Fuzzy Possibilistic C-Means algorithms that viewed the accuracy comparison of both methods used the Modified Partition Coefficient (MPC) cluster validity. The clustering process was carried out five times by the number of clusters ranging from 3 to 7 clusters. The results of the study showed the Fuzzy C-Means method is a better method compared to the Fuzzy Possibilistic C-Means method in clustering data tweets, with the number of clusters formed is 4. The content type formed is related to promo, discount, cashback, prize quizzes, and event promotions organized by Tokopedia. Content with the highest average number of retweets and likes is about automotive deals, sports tools, and merchandise offerings. So, that PT Tokopedia can use this content type as a tool for advertising on Twitter because it gets more likes by followers of @tokopedia.Keywords: Data Tweets, Clustering, Fuzzy C-Means, Fuzzy Possibilistics C-Means, Modified Partition Coefficient.

Analysis of container terminal accidents using fuzzy c-means

Maritime transportation is the key element of global trade in recent years. Moreover, container shipping companies aim to complete door-to-door cargo operation with lower costs and increased speed. Determination of delays and possible accidents play vital role on cargo transportation process in a container terminal. In this scope, 65 container terminal accident reports are evaluated with Fuzzy C-means method to cluster within scope of root cause analysis. Three, four, and five number of clusters are created to determine possible causes of accidents. Two container terminals located in Turkey have diverse cargo operation capacity such as mobile and quay cranes. Accident cases are grouped in the view of relation of operations as cargo and port. Port related accidents are found mainly as related with maintenance processes. and occurred in weekdays. On the other hand, cargo related accidents occurred in weekends and during night shifts. In this scope, the results revealed that Fuzzy C-Means method could be used to analyse container terminal accidents.

Multiscale Laplacian graph kernel features combined with tree deep convolutional neural network for the detection of ECG arrhythmia

In this manuscript, Multiscale Laplacian graph kernel features combined with Tree Deep Convolutional Neural Network (MLGK-TDCNN) is proposed for the detection of Electrocardiogram (ECG) arrhythmia. Here, the input ECG signals are taken from two datasets: (i) MIT-BIH AF database (AFDB) (ii) MIT-BIH arrhythmia database (MBDB). These datasets are fully unbalanced dataset, and these datasets are balanced using Improved fuzzy c-means method for unbalanced dataset. Moreover, the de-noised ECG signals are given to the MLGK. The proposed MLGK is to combine the Multiscale kernel features from the Preprocessed ECG signals. The combined Multiscale kernel features given to the TDCNN classifier for the detection of AF with raw normal sinus rhythm (NSR). The proposed approach is activated in MATLAB platform, then the efficiency is analyzed with existing approaches. The experimental outcomes demonstrate that the proposed FFREWT-MLGK-TDCNN approach is compared with two databases. From the analysis, the accuracy of AFDB shows 9.40%, 16.44% and 23.20% better than the existing approaches, the accuracy of MBDB shows 14.67%, 21.42% and 7.54% better than the existing approaches, like novel intelligent approach depending on multi-scale convolution kernel (MCK) and Squeeze-and-Excitation network (SENet) for AF detection, automatic arrhythmia classification strategy using the optimization-based deep convolutional neural network (CNN-BaROA), deep learning method for classifying arrhythmia by using 2-second segments of 2D recurrence plot images of ECG signals (2D-CNN) respectively.

Fuzzy-Weighted Echo State Networks

A novel echo state network (ESN), referred to as a fuzzy-weighted echo state network (FWESN), is proposed by using the structural information of data sets to improve the performance of the classical ESN. The information is incorporated into the classical ESN via the concept of Takagi–Sugeno (TS) models/rules. We employ the fuzzy c-mean clustering method to extract the information based on the given data set. The antecedent part of the TS model is determined by the information. Then, we obtain new fuzzy rules by replacing the affine models in the consequent part of each TS rule with a classical ESN. Consequently, the output of the proposed FWESN is calculated through inferring these new fuzzy rules by a fuzzy-weighted mechanism. The corresponding reservoir is consisted of the sub-reservoirs of the new fuzzy rules. Furthermore, we prove that the FWESN has an echo state property by setting the largest spectrum radium of all the internal weight matrices in the sub-reservoirs less than one. Finally, a nonlinear dynamic system and five nonlinear time series are employed to validate the FWESN.

A Study for Plausible Third Wave of COVID-19 in India through Fuzzy Time Series Modelling Based on Particle Swarm Optimization and Fuzzy c-Means

The outbreak of COVID-19 has become a global pandemic as announced by World Health Organisation. As India has already met the two waves, named first and second wave, it is assumed that COVID-19 will again strike in India in the form of third wave. The peak during the upcoming third wave and determination of the approximated maximum number of COVID-19 infected cases and deaths at a particular day becomes crucial for India. To determine the peak of infectious curve, this article proposed a hybrid fuzzy time series forecasting model based on particle swarm optimization and fuzzy c-mean technique, named as fuzzy time series particle swarm optimization extended fuzzy c-mean technique. The proposed model works in two phases. In phase-I, particle swarm optimization extended fuzzy c-mean method is used to form initial intervals with the help of centroids, while in phase-II, these intervals are updated to form subintervals. In the present article, a fitness function is developed for particle swarm optimization to increase its convergence speed and basic fuzzy c-mean is extended by using an exponential function to tolerate the effect of outliers, named as extended fuzzy c-mean technique. The effectiveness of the proposed model has been tested based on mean square error and root mean square error on first and second wave COVID-19 data, and the obtained results are very close to the existing data of COVID-19 with less error rate. Thus, the proposed model is suitable to forecast a better approximation value of COVID-19 infected cases and deaths in India during the upcoming third wave. This study demonstrates that third wave of COVID-19 could occur in India, while also illustrating that it is unlikely for any such resurgence to be as large as the second wave. The proposed model predicts that the peak of third wave will occur approximately after 40–70 days from the mid of December. Furthermore, the impact of vaccination on infected cases and deaths during the upcoming third wave in India is also studied. With the implementation of the vaccine on the Indian people, the peak of COVID-19 infected during third wave will be shifted in forward direction. On the basis of the proposed model, government authorities will be enabling to know expected required resources such as hospital patient beds, ICU beds, and oxygen concentrators during the upcoming outspread of COVID-19 like disease in future.

Electric vehicle user classification and value discovery based on charging big data

With the rapid development of electric vehicles (EVs) in recent years, it is important to understand the varied EV users for EV sector business innovation. Therefore, identifying different types of EV users and implementing differentiated marketing strategies can assist charging service enterprises in improving profitability and user loyalty. Recency, frequency and monetary (RFM) model is an important data mining method that has important practical applications in customer relationship management and direct marketing fields. To classify EV users, an integrated approach incorporating an extended RFM model, a two-stage clustering method, and the Entropy Weight Method is proposed in this study. Analysis results demonstrate that 7426 EV users are divided into six groups, namely “high value users”, “key users to maintain”, “key users to develop”, “potential users”, “new users” and “lost users”. To estimate the performances of the proposed approach, the traditional cluster algorithm and fuzzy c-means method are compared with the improved entropy-cluster algorithm by using the intraclass method. The results indicate that the proposed approach is more robust than other methods. Finally, we develop related marketing strategies for each group of EV users to assist charging service enterprises in improving their marketing effectiveness and financial performance.

Nonlinear model-based cardiac arrhythmia diagnosis using the optimization-based inverse problem solution.

This study investigates a nonlinear model-based feature extraction approach for the accurate classification of four types of heartbeats. The features are the morphological parameters of ECG signal derived from the nonlinear ECG model using an optimization-based inverse problem solution. In the model-based methods, high feature extraction time is a crucial issue. In order to reduce the feature extraction time, a new structure was employed in the optimization algorithms. Using the proposed structure has considerably increased the speed of feature extraction. In the following, the effectiveness of two types of optimization methods (genetic algorithm and particle swarm optimization) and the McSharry ECG model has been studied and compared in terms of speed and accuracy of diagnosis. In the classification section, the adaptive neuro-fuzzy inference system and fuzzy c-mean clustering methods, along with the principal component analysis data reduction method, have been utilized. The obtained results reveal that using an adaptive neuro-fuzzy inference system with data obtained from particle swarm optimization will have the shortest process time and the best diagnosis, with a mean accuracy of 99% and a mean sensitivity of 99.11%.

Choriocapillaris Changes in Myopic Macular Degeneration.

PurposeMyopic macular degeneration (MMD) can cause irreversible vision loss. Thinner choroid is associated with increased MMD severity. This cross-sectional study analyzed choriocapillaris (CC) alterations in MMD.MethodsAxial length (AL), best-corrected visual acuity (BCVA), fundus photography, and swept-source optical coherence tomography angiography (SS-OCTA) were assessed in controls and high myopes (spherical equivalent ≤ −6 diopters). Myopic patients with grade 2 MMD (macular diffuse chorioretinal atrophy [MDCA]), high axial myopia (AL ≥ 26.5 mm), and BCVA ≥ 20/40 were compared with controls without MMD. CC mean thickness was measured from 3 × 3-mm SS-OCTA scans by identifying CC peaks in A-scan intensity profiles. CC flow deficit percent (CC FD%) was quantified using a fuzzy C-mean local thresholding method on en face OCTA images. Multivariate regressions compared CC thickness and CC FD% between myopic patients and controls, correcting for age and other confounders.ResultsSixteen eyes with MDCA (AL, 26.96–33.93 mm; ages, 40–78 years) were compared with 51 control eyes (AL, 21.65–25.84 mm; ages, 19–88 years). CC thickness in patients with MDCA was 66% lower than that in controls (5.23 ± 0.68 µm [mean ± SD] vs. 15.46 ± 1.82 µm; P < 0.001). CC FD% in patients with MDCA was 237% greater than in controls (26.5 ± 4.3 vs. 11.2 ± 4.6; P < 0.001).ConclusionsPatients with MDCA with good visual acuity had thinner CC and increased CC FD%, or reduced CC flow, compared with controls. Patients with grade 2 MMD and good visual acuity demonstrated significant choriocapillaris alterations, suggesting that choriocapillaris perfusion defects contribute to the pathogenesis of MMD.Translational RelevanceGiven the potential vascular etiology for MMD, current research about revascularization of ischemic retina likely has implications for the treatment of MMD.

Optimization of fuzzy c-means (FCM) clustering in cytology image segmentation using the gray wolf algorithm

BackgroundImage segmentation is considered an important step in image processing. Fuzzy c-means clustering is one of the common methods of image segmentation. However, this method suffers from drawbacks, such as sensitivity to initial values, entrapment in local optima, and the inability to distinguish objects with similar color intensity. This paper proposes the hybrid Fuzzy c-means clustering and Gray wolf optimization for image segmentation to overcome the shortcomings of Fuzzy c-means clustering. The Gray wolf optimization has a high exploration capability in finding the best solution to the problem, which prevents the entrapment of the algorithm in local optima. In this study, breast cytology images were used to validate the methods, and the results of the proposed method were compared to those of c-means clustering.ResultsFCMGWO has performed better than FCM in separating the nucleus from the other dark objects in the cell. The clustering was validated using Vpc, Vpe, Davies-Bouldin, and Calinski Harabasz criteria. The FCM and FCMGWO methods have a significant difference with respect to the Vpc and Vpe indexes. However, there is no significant difference between the performances of the two clustering methods with respect to the Calinski-Harabasz and Davies-Bouldin indices. The results indicate the better efficacy of the proposed method.ConclusionsThe hybrid FCMGWO algorithm distinguishes the cells better in images with less detail than in images with high detail. However, FCM exhibits unacceptable performance in both low- and high-detail images.

A hybrid spatial model based on identified conditions for 3D pore pressure estimation

Accurate spatial modeling of pore pressure helps to provide extensive and continuous information regarding the insightful downhole geological environment for safe drilling plans, performance analysis, and efficient reservoir modeling. Considering the varying drilling depth and pore pressure of study wells, a hybrid spatial modeling method is developed to estimate the pore pressure of an area. The method comprises three parts. In the first part, an appropriate number of pressure conditions is determined adaptively based on quantization error modeling (QEM), and similar pressure conditions are identified based on fuzzy c-means (FCM) method. In the second part, a kriging interpolation (Kriging) algorithm is introduced to produce profiles of different pressure subspaces. Afterward, random forest (RF) submodels are built separately according to the pressure profiles. In the last part, all of these submodels coalesced into one whole model. Verification of the proposed method is carried out on real case studies using field data from Songliao basin. The results show that the proposed method achieves the best prediction accuracy than the other five well-known methods for six wells, especially for well XS3, the mean absolute error and mean absolute percentage error of the proposed method achieve 0.717 and 6.063, which are at least 31% less than those of other five well-known methods.

Penggerombolan Kabupaten/Kota di Indonesia Berdasarkan Indikator Indeks Pembangunan Manusia Menggunakan Metode K-Means dan Fuzzy C-Means

The achievement of the human development index in Indonesia differs between regions with striking gaps occurring in the western and eastern parts of Indonesia. This difference in achievement can be seen more clearly by grouping regencies/municipalities in Indonesia based on the four indicators of the human development index. With this aim, this study uses the k-means and fuzzy c-means methods to determine the optimal cluster size with two distance approaches, namely the Euclidean and Manhattan distances on the human development index indicators data in 2020. In addition, this study also seeks to identify the distribution of regencies/municipalities based on the characteristics of the human development index indicators in the clustering result. The result is that the best distance measure is Euclidean distance with optimal cluster size is four for k-means and six for fuzzy c-means. In addition, the clustering results obtained by the k-means method are more optimal than the fuzzy c-means because the evaluation value is better. In general, the four clusters formed were in accordance with the grouping carried out by BPS with the percentage of conformity reaching 66,54%. In summary, most regencies/municipalities on the Island of Sumatera, Java, Borneo and Sulawesi have higher life expectancy and percapita expenditure than many regencies/municipalities in the Nusa Tenggara Islands (besides Bali), Moluccas and Papua. Very high achievement for each HDI indicators is dominated by the capital city of each province with unfavorable conditions occurring in most regencies/municipalities in Papua Province.

On efficient model selection for sparse hard and fuzzy center-based clustering algorithms

The class of center-based clustering algorithms offers methods to efficiently identify clusters in data sets, making them applicable to larger data sets. While a data set may contain several features, not all of them may be equally informative or helpful towards cluster detection. Therefore, sparse center-based clustering methods offer a way to select only those features that may be useful in identifying the clusters present in a data set. However, to automatically determine the degree to which features should be selected, these methods use the Permutation Method which involves generating and clustering multiple randomly permuted data sets, leading to much higher computation costs. In this paper, we propose an improved approach towards model selection for sparse clustering by using expressions of Bayesian Information Criterion (BIC) derived for the center-based clustering methods of k-Means and Fuzzy c-Means. The derived expressions of BIC require significantly lower computation costs, yet allow us to compare and select a suitable sparse clustering among several possible sparse partitions that may have selected different subsets of features. Experiments on synthetic and real-world data sets show that using BIC for model selection leads to remarkable improvements in the identification of sparse clusterings for both Sparse k-Means and Sparse Fuzzy c-Means.

Implementation of Fuzzy C-Means to Determine Student Satisfaction Levels in Online Learning

Online learning is the best solution during the COVID-19 pandemic, causing a difference in the learning process that lecturers usually do face-to-face with students. Changes in the learning process are expected to be implemented effectively and efficiently. The purpose of this study is to determine how the level of student satisfaction in online learning and produce a system that can cluster the level of student satisfaction using the Fuzzy C-Means method. The variables used in this study have 3 parameters, namely the teaching and learning process, lecturer competence, facilities, and infrastructure. The stages carried out in this study were collecting questionnaire data filled out by 500 students of Informatics Engineering, Malikussaleh University, the data obtained was processed based on the average student answers. Based on the data that has been obtained will form 2 clusters, namely satisfied and dissatisfied in online learning using the Fuzzy C-Means method. The results of clustering of 500 data obtained that 303 students felt "satisfied" and as many as 197 students felt "not satisfied" in the online learning carried out, for cluster 1 (satisfied students) with an average value of the teaching and learning process 7,389, lecturer competence 7,863 and facilities and infrastructure 7,8883. Cluster 2 (students are not satisfied) with the average value of teaching and learning process 4,466, lecturer competence 5,069, and the value of facilities and infrastructure 5,424. The contribution of this research is to help the manager of the Department of Information Engineering, Malikussaleh University to evaluate and improve the online learning process.

Analysis of Rice Field Cluster in Indonesia as an Evaluation of Food Production Availability Using Fuzzy C-Means

Rice fields area in Indonesia is getting narrower every year with the rampant construction of housing and buildings. It results in lower availability of food production hence to meet the needs we have to import rice from other countries. By clustering rice fields, it can be used as an evaluation material to increase food production in Indonesia so that the need for rice imports can be minimized. The method used in the grouping of Rice Fields is the Fuzzy C-means method, implementation of the Knime Tool with data training and testing. The Fuzzy C-Means program produces three data groups/clusters, namely wide, moderate, and narrow rice fields. The results of the clustering show that the most potential areas for food production from rice fields are East Java, Central Java, and West Java.

Comparison of clustering in tuberculosis using fuzzy c-means and k-means methods

Tuberculosis (TB) is a health problem that has yet to be resolved in Indonesia. Based on WHO data, in 2021 Indonesia will still be in the third rank of the highest TB cases in the world. This study aims to determine how many groups of TB patients are based on age, gender, HIV status, history of diabetes mellitus, chest X-ray, and the results of the Molecular Rapid Test (TCM). The data used in this study were 985 from 2017 to 2020. The method used in this research is K-Nearest Neighbor (KNN) in carrying out the imputation process, as well as comparing the k-means and Fuzzy C-Means (FCM) methods in classifying TB data. Before doing the grouping, the data cleaning process is carried out by an imputation process which is useful for filling in the missing data in this case, using the KNN method. To produce maximum results of data grouping or clustering, it is necessary to determine the right number of clusters. For this reason, this study tries to compare the elbow, silhouette coefficient, and Davies Bouldin Index (DBI) methods. The application of the KNN method in the data imputation process in this study is to use k=5. The application of the K-Means algorithm is to form groups of TB patients based on six features. Determination of the optimal number of clusters using the K-means and FCM methods shows the optimal number of clusters, namely K = 2 but with different values. The results of the clustering test using the elbow method with the K-means and FCM methods are 93288.49. The DBI value for the K-means and FCM methods is 0.4937. Meanwhile, the clustering trial with the silhouette coefficient on K-means yields a value of 0.6318 which is better than the FCM which produces a value of 0.6321. This shows that the results of clustering k-means with silhouette coefficients produce better cluster quality because they have a lower silhouette coefficient value than FCM.