K-medoids Algorithm Research Articles

Mother-reported pain experience between ages 7 and 10: A prospective study in a population-based birth cohort.

Trajectory studies suggest considerable stability of persistent or recurrent pain in adolescence. This points to the first decade of life as an important aetiologic window for shaping future pain, where the potential for prevention may be optimised. We aimed to quantify changes in mother-reported pain experience in children between ages 7 and 10 and describe clusters of different pain experiences defined by complementary pain features. We conducted a prospective study using data from 4036 Generation XXI birth cohort participants recruited in 2005-06. Pain history was reported by mothers at ages 7 and 10 using the Luebeck pain screening questionnaire. We tracked changes in six pain features over time using relative risks (RRs) and their 95% confidence intervals (95% CIs). Clusters were obtained using the k-medoids algorithm. The risk of severe pain at age 10 increased with increasing severity at age 7, with RRs ranging from 2.18 (95% CI 1.90, 2.50) for multisite to 4.43 (95% CI 3.19, 6.15) for high frequency pain at age 7. A majority of children (59.4%) had transient or no pain but two clusters included children with stable recurrent pain (n=404, 10.2% of the sample). One of those (n=177) was characterised by higher probabilities of multisite pain (74.6% and 66.7% at ages 7 and 10, respectively), with psychosocial triggers/contexts (59.3% and 61.0%) and daily-living restrictions (72.2% and 84.6%). Most children in that cluster (58.3%) also self-reported recent pain at age 10 and had more frequent family history of chronic pain (60.5%). All pain features assessed tracked with a positive gradient between ages 7 and 10, arguing for the significance of the first decade of life in the escalation of the pain experience. Multisite pain and psychosocial attributions appeared to be early markers of more adverse pain experiences.

Multivariate time series clustering and forecasting for building energy analysis: Application to weather data quality control

In recent years, several tools for building energy analysis and simulation have been developed to assist in increasing building energy performance, harvesting its computing capabilities for a reliable and accurate energy performance prediction. To perform this analysis, energy tools typically require crucial data regarding the building's surrounding environment, which is acquired from neighbouring weather stations. However, these stations often experience hardware malfunctions, resulting in either erroneous or missing data. Traditionally, these values are rectified through empirical and geostatistical methods, which, while reflecting several decades of practice, may prove to be inadequate when considering a purely data-driven approach. To this end, the present study introduces a machine learning methodology proposing the application of regression algorithms to rectify the erroneous values in datasets, and the clustering of weather stations, based on their recorded climatic conditions, to enhance the regression models. A shape-based approach for clustering time series of different climatic parameters and weather stations is pursued, using the k-medoids algorithm alongside dynamic time warping as the similarity measure. Both Artificial Neural Networks (ANN) and Support Vector Regression (SVR) models are evaluated as exemplary regression algorithms, with different sets of predictors. Mean Squared Error is used as the performance metric. A data set of different climatic parameters from southeastern Brazil was used, with air temperature being chosen as the response variable, given its importance in energy consumption. Results indicate that a machine learning approach to the problem is indeed viable. ANN slightly outperforms SVR in the prediction of the studied weather variable.

Breakthrough Cancer Pain Clinical Features and Differential Opioids Response: A Machine Learning Approach in Patients With Cancer From the IOPS-MS Study.

PURPOSEA large proportion of patients with cancer suffer from breakthrough cancer pain (BTcP). Several unmet clinical needs concerning BTcP treatment, such as optimal opioid dosages, are being investigated. In this analysis the hypothesis, we explore with an unsupervised learning algorithm whether distinct subtypes of BTcP exist and whether they can provide new insights into clinical practice.METHODSPartitioning around a k-medoids algorithm on a large data set of patients with BTcP, previously collected by the Italian Oncologic Pain Survey group, was used to identify possible subgroups of BTcP. Resulting clusters were analyzed in terms of BTcP therapy satisfaction, clinical features, and use of basal pain and rapid-onset opioids. Opioid dosages were converted to a unique scale and the BTcP opioids-to-basal pain opioids ratio was calculated for each patient. We used polynomial logistic regression to catch nonlinear relationships between therapy satisfaction and opioid use.RESULTSOur algorithm identified 12 distinct BTcP clusters. Optimal BTcP opioids-to-basal pain opioids ratios differed across the clusters, ranging from 15% to 50%. The majority of clusters were linked to a peculiar association of certain drugs with therapy satisfaction or dissatisfaction. A free online tool was created for new patients’ cluster computation to validate these clusters in future studies and provide handy indications for personalized BTcP therapy.CONCLUSIONThis work proposes a classification for BTcP and identifies subgroups of patients with unique efficacy of different pain medications. This work supports the theory that the optimal dose of BTcP opioids depends on the dose of basal opioids and identifies novel values that are possibly useful for future trials. These results will allow us to target BTcP therapy on the basis of patient characteristics and to define a precision medicine strategy also for supportive care.

NLP Algorithms Endowed f or Automatic Extraction of Information from Unstructured Free Text Reports of Radiology Monarchy

Natural Language Processing (NLP) Algorithms are the key factors for automatic information extraction form the unstructured free-text radiology reports .To extract clinically important findings and recommendations, various NLP algorithms are used . A rule-based NLP system is used in most of the automated IE applications in medical domain; whereas some applications are using Random Forest classifier, PageRank Algorithm, clustering algorithm, Conditional Random Fields (CRF) algorithm, and deep learning-based approaches. Some papers found with methods used for IE like, Support Vector Machines (SVMs), linear-chain conditional random fields (LC-CRFs), k-means or k-medoids algorithm ,Affinity Propagation (AP) clustering algorithm , supervised machine learning algorithm and many more. Thus through this survey we can say that, NLP methods used to extract information ,brings new insights into already known clinical evidences. It also helps to identify previously unknown treatment and causal relations between biomedical entities .Therefore NLP algorithms has empowered Radiology monarchy.

The K-Medoids Clustering Method for Learning Applications during the COVID-19 Pandemic

A disease that is currently widespread today is caused by the spread of the coronavirus disease or what is commonly called COVID 19. This virus is very dangerous to health because it attacks organs in the human body from various sources, either from the air or direct touch. With the existence of COVID 19, it has an impact on all countries, especially the State of Indonesia, which consists of various islands, which are also affected by the COVID 19 virus. So that the central government takes a policy to carry out social distancing to every one to break the chain of spreading this virus, with this social distancing it has an impact on all activities that occur every day. As an impact on the learning process that usually takes place in class, it turns into online learning that uses several supporting applications in the learning process during the COVID 19 pandemic. With online learning from various applications, it attracts researchers to research with the K-Medoid Clustering Algorithm in using applications during the pandemic COVID 19.

An Improvement of K-Medoids Clustering Algorithm Based on Fixed Point Iteration

The process of K-medoids algorithm is that it first selects data randomly as initial centers to form initial clusters. Then, based on PAM (partitioning around medoids) algorithm, centers will be sequential replaced by all the remaining data to find a result has the best inherent convergence. Since PAM algorithm is an iterative ergodic strategy, when the data size or the number of clusters are huge, its expensive computational overhead will hinder its feasibility. The authors use the fixed-point iteration to search the optimal clustering centers and build a FPK-medoids (fixed point-based K-medoids) algorithm. By constructing fixed point equations for each cluster, the problem of searching optimal centers is converted into the solving of equation set in parallel. The experiment is carried on six standard datasets, and the result shows that the clustering efficiency of proposed algorithm is significantly improved compared with the conventional algorithm. In addition, the clustering quality will be markedly enhanced in handling problems with large-scale datasets or a large number of clusters.

Research on Optimal Load Scheduling for Multiple Scenarios

In recent years, the operation of power grid is under great pressure due to more and more distributed resources. This paper establishes three kinds of load models, namely the electric water heater(EWH), the electric vehicle(EV) and the energy storage(ES). The response cost, speed, capacity and duration are taken as the four characteristic elements to cluster the demand side loads. According to the scenario characteristics, the factor weights under different scenarios are determined. Furthermore, the particle swarm optimization (PSO) algorithm utilized to optimize the scheduling of demand-side resources selected by K-medoids clustering algorithm. Simulation results indicate that, by taking advantage of the complementarity of the time-domain and functional characteristics of multiple loads, the consumption needs of different scenarios can be satisfied.

Sentence Embedding Based Semantic Clustering Approach for Discussion Thread Summarization

Huge data on the web come from discussion forums, which contain millions of threads. Discussion threads are a valuable source of knowledge for Internet users, as they have information about numerous topics. The discussion thread related to single topic comprises a huge number of reply posts, which makes it hard for the forum users to scan all the replies and determine the most relevant replies in the thread. At the same time, it is also hard for the forum users to manually summarize the bulk of reply posts in order to get the gist of discussion thread. Thus, automatically extracting the most relevant replies from discussion thread and combining them to form a summary are a challenging task. With this motivation behind, this study has proposed a sentence embedding based clustering approach for discussion thread summarization. The proposed approach works in the following fashion: At first, word2vec model is employed to represent reply sentences in the discussion thread through sentence embeddings/sentence vectors. Next, K-medoid clustering algorithm is applied to group semantically similar reply sentences in order to reduce the overlapping reply sentences. Finally, different quality text features are utilized to rank the reply sentences in different clusters, and then the high-ranked reply sentences are picked out from all clusters to form the thread summary. Two standard forum datasets are used to assess the effectiveness of the suggested approach. Empirical results confirm that the proposed sentence based clustering approach performed superior in comparison to other summarization methods in the context of mean precision, recall, and F-measure.

Performance Assessment of University Lecturers: A Data Mining Approach

A lecturer with a good performance has a positive impact on the quality of teaching and learning. The said quality includes the delivery of teaching materials, learning methods, and ultimately the academic results of students. Performance of lecturers contributes significantly to the quality of research and community service which in turn improves the quality of teaching materials. It is desirable, therefore, to have a method to measure the performance of lecturers in carrying out the Tri Dharma (or the three responsibility) activities, which consist of teaching and learning process, research, and community service activities, including publications at both national and international level. This study seeks to measure the performance of lecturers and cluster them into three categories, namely "satisfactory", "good", and "poor". Data were taken from academic works of nursing study program lecturers in conducting academic activities. Clustering process is carried out using two machine learning approaches, which is K-Means and K-Medoids algorithms. Evaluation of the clustering results suggests that K-Medoids algorithm performs better compared to using K-Means. DBI score for clustering techniques using K-Means is -0.417 while the score for K-Medoids is -0.652. The significant difference in the score shows that K-Medoids algorithm works better in determining the performance of lecturers in carrying out Tri Dharma activities.

A novel landslide susceptibility mapping portrayed by OA-HD and K-medoids clustering algorithms

Because of the strong dependence on the values for the input parameters and the cluster shape, as well as the difficulties in quantifying the precipitation in constructing landslide susceptibility maps by employing existing clustering algorithms, we propose a novel method based on an Ordering Points to Identify the Clustering Structure (OPTICS) algorithm using the Hausdorff distance (OA-HD). The OA-HD algorithm distributes mapping units into many subclasses with similar characteristic values for topography and geology. To obtain more optimal subclasses, the HD was adopted to quantify precipitation. The K-medoids algorithm grouped these subclasses into five susceptibility levels according to the values of landslide density in each subclass. Applying the innovative integrated algorithms to the study area significantly improves the landslide susceptibility assessment, especially in a large study area. The method suggests new insights for better assessing landslide susceptibility in a large study area.

Improved K-medoids algorithm-based clustering analysis for handle driving force in automotive manual sliding door closing process

Handle driving forces are the input of the automotive sliding door dynamic system and play an important role for ensuring a smooth closing process during manual sliding door mechanism design. It is important to provide a reliable and accurate input for the manual sliding door mechanism during the design and analysis stage. This paper aims to present an improved K-medoids clustering algorithm to investigate the characteristic of handle driving forces in manually closing an automotive sliding door based on experimental data. The improved K-medoids clustering algorithm includes two stages: observation-based clustering stage and traditional K-medoids clustering stage. In all, 134 subjects have been recruited to manually close the sliding door in the lab and the handle driving force data are collected and processed. The handle driving forces are described in the sliding door coordinate system (XYZ) fixed on the door. This study mainly focuses on the X direction force component clustering analysis. The first stage of the improved algorithm classifies the X direction force components into three clusters based on force curve shapes. Then, each of the above identified three clusters is clustered with the traditional K-medoids clustering algorithm. Results show that the X direction force component has three different shapes: Shape 1—only one crest in the curve, Shape 2—two crests in the curve, and Shape 3—one crest and one trough in the curve. The forces with three different shapes are finally divided into six clusters and the amplitude and time duration are similar for X direction forces within the same cluster and are different in the different clusters. The medoids of these clusters are the mined representative prototypes. Compared to the pure traditional K-medoids algorithm, the improved algorithm can provide much better results that give insights on subjects’ door closing behaviors.

Clustering and Characterization of the Lactation Curves of Dairy Cows Using K-Medoids Clustering Algorithm.

Simple SummaryA lactation curve (LC) provides valuable insights in planning appropriate management strategies related to health, nutrition, and breeding in dairy cows. A clustering based approach on LC patterns analysis is presented. The k-medoids algorithm is adopted for the clustering. This approach generates several clusters which have similar milking characteristics of total milk yield, peak milk yield, and days in milk at peak yield. The LCs of some groups represent characteristics of atypical milking patterns which are not considered much in previous approaches, whereas LCs of the other groups show the typical LC patterns similar to the results of previous methods. This approach could be used as a tool to manage an abnormal herd of cows.The aim of the study was to group the lactation curve (LC) of Holstein cows in several clusters based on their milking characteristics and to investigate physiological differences among the clusters. Milking data of 330 lactations which have a milk yield per day during entire lactation period were used. The data were obtained by refinement from 1332 lactations from 724 cows collected from commercial farms. Based on the similarity measures, clustering was performed using the k-medoids algorithm; the number of clusters was determined to be six, following the elbow method. Significant differences on parity, peak milk yield, DIM at peak milk yield, and average and total milk yield (p < 0.01) were observed among the clusters. Four clusters, which include 82% of data, show typical LC patterns. The other two clusters represent atypical patterns. Comparing to the LCs generated from the previous models, Wood, Wilmink and Dijsktra, it is observed that the prediction errors in the atypical patterns of the two clusters are much larger than those of the other four cases of typical patterns. The presented model can be used as a tool to refine characterization on the typical LC patterns, excluding atypical patterns as exceptional cases.

Cauliflower Seedling Recognition Using Stereo Vision and K-Medoids Algorithm under Weedy Conditions

The natural field environment is unstructured where the illumination conditions are uneven, the soil background is complex and changeable. Also, the improper of field management is likely to cause weeds whose colour and shape are similar to crops. Those factors make it hard to identify crops quickly and stably by traditional 2D image processing technologies. So the cauliflower seedling recognition method using binocular stereo vision and K-medoids was proposed in this paper. For the disparity map the Sum of Squared Differences (SSD) algorithm was used in the step of stereo matching. For the cauliflower seedling recognition the K-medoids algorithm was applied in the step of clustering. The experiment results showed that the average operation time of the SSD algorithm was 190ms. The real-time performance and matching effect of the SSD algorithm were superior to Belief Propagation algorithm (BP) and Dynamic Programming algorithm (DP). The average running time of the proposed method for processing the left image and its corresponding right images was 379ms, which meet the real-time requirements. The correct recognition rate of the method proposed was 98.75% of 240 pairs of cauliflower seedling images.

Cluster analysis application to identify groups of individuals with high health expenditures

We compare and demonstrate the effectiveness of two clustering methods with the main purpose of identifying characteristic profiles of high utilizers of health care. In this work, we use three sets of mutually independent longitudinal data that are nationally representative of the US adult working-age civilian non-institutionalized population. We compare k-means, a commonly used clustering method, with a k-medoids algorithm called Partitioning Around Medoids. We use one cohort of data to create clusters based on similar characteristics of individuals for both clustering methods. We examine these characteristic compositions of the highest three average total expenditure clusters from this cohort. We also examine the health expenditure distributions for this cohort over the following two years. We validate the approach by applying the centers of the clusters to two other cohorts of similar data. We form clusters based on demographic, economic, and health-related characteristics that are commonly used in studies of health care utilization. We demonstrate the consistency of our results across the three cohorts of data and across different types of health expenditures, such as office-based/outpatient and drug. Clusters can be formed with other more homogeneous data, such as Medicaid, Medicare, employer sponsored insurance, or individual private plans issued under the Affordable Care Act. This approach can be used to follow similar groups over time for other types of health outcomes.

Implementasi Algoritma K-Medoids untuk Pengelompokkan Sebaran Mahasiswa Baru

The existence of new students at a tertiary institution is a routine activity every year in a tertiary institution and can also see the sustainability of the tertiary institution. The variety of regional origin of new students makes the party from the university want to see the distribution of new students based on the origin of the school and its place of residence. STIKOM Tunas Bangsa is one of the tertiary institutions in Pematangsiantar. It aims to promote the university. K-Medoids is able to group data on the distribution of new students in STIKOM Tunas Bangsa Pematangsiantar. The clusters produced in this study are of three clusters. The validity used in this study is the validity of Silhoutte Coefficient. The validity value generated in the K-Medoids algorithm produces a validity value of -116.47 by assuming that if the non-medoids value produced S

Algoritma K-Medoids Untuk Menentukan Calon Mahasiswa Yang Layak Mendapatkan Beasiswa Bidikmisi di Universitas Budi Darma

Bidikmisi scholarship is a government program to help prospective new students who are academically capable and economically incapable. The Bidikmisi scholarship is a form of tuition assistance and living expenses. Starting in 2018 Budi Darma University began accepting new students through the bidikmisi scholarship path, admission of new students through the bidikmisi path must meet the requirements set by the government. Determination of whether or not a prospective new student is a recipient of bidikmisi based on report cards, school performance, results of selection tests and interviews. During this time the organizers and managers of bidikmisi at Budi Darma University have had difficulty determining prospective students who are truly eligible to receive bidikmisi scholarships in addition to the very limited quota and the large number of prospective students receiving bidikmisi scholarships as well as the value of each criteria for prospective students receiving bidikmisi which is almost the same or similar to one another. To make it easier to determine prospective students receiving the Bidikmisi scholarship, the K-Medoids algorithm is used. K-Medoids algorithm is one of the algorithms in data mining to group data based on the closest criteria value

Scheduling optimization model for integrated energy systems considering conditional value-at-risk

The integrated energy system (IES) promotes the integration of renewable energy and improves energy efficiency, but it also brings uncertainty to the system. The uncertainty of the output of renewable energy may lead to imbalance of supply and demand, which brings the risk of cost increase to the economic dispatch. This paper mainly studies IES with energy hub as the research object, and studies the IES modeling and the optimal scheduling problem considering Conditional Value-at-Risk (CVaR). Firstly, the mathematical model of energy hub is established. Next, the network model of power system and natural gas system is established and the energy hub is reasonably connected to them. Then, for analyzing the uncertainty of wind power output, this paper develops scenarios using Latin hypercube sampling method and uses K-medoids clustering algorithm to reduce the scenarios. After defining the risk cost under uncertainty, CVaR is introduced into the objective function of IES optimal dispatch to describe the risk which operating personnel faces. Based on CVaR, the IES scheduling model is established and linearized into a mixed integer linear programming problem to be solved. Finally, numerical case studies verify that the IES scheduling model can reduce the risks with least costs and determine whether the system needs to avoid risks.

A New Representative Sampling Method for Series Size Rock Joint Surfaces

The greatest variability in both shear strength and roughness exists for joint samples with smaller size, which underscores the necessity of performing representative sampling. This study aims to provide a representative sampling method for series size joint surfaces. The progressive coverage statistical method is introduced to provide the sufficient sample capacity for series sampling sizes by setting different propulsion spaces. The statistical law of the joint surface morphology at different sampling sizes is measured by the 3D roughness parameter with {{theta }}_{max }^{ast }/({C}+1). Through an application in nine natural large-scale rock joints, nine consecutive sampling sizes from 100 mm × 100 mm to 900 mm × 900 mm are selected and 121 samples are successfully acquired from each sampling size. According to the frequency distribution of roughness statistics, a new sampling method combining the layering principle and K-medoids clustering algorithm is proposed to screen representative joint samples for each sampling size. The sampling results that meet the test accuracy requirements suggest the possibility of realizing an intelligent sampling method. In addition, the representative of the interlayer cluster center is validated. Finally, the comparison results with the traditional stratified sampling method prove that the proposed method has better stability.

Research on Phase Combination and Signal Timing Based on Improved K-Medoids Algorithm for Intersection Signal Control

Aiming at the problem of intersection signal control, a method of traffic phase combination and signal timing optimization based on the improved K-medoids algorithm is proposed. Firstly, the improvement of the traditional K-medoids algorithm embodies in two aspects, namely, the selection of the initial medoids and the parameter k, which will be applied to the cluster analysis of historical saturation data. The algorithm determines the initial medoids based on a set of probabilities calculated from the distance and determines the number of clusters k based on an exponential function, weight adjustment, and elbow ideas. Secondly, a phase combination model is established based on the saturation and green split data, and the signal timing is optimized through a bilevel programming model. Finally, the algorithm is evaluated over a certain intersection in Hangzhou, and results show that this algorithm can reduce the average vehicle delay and queue length and improve the traffic capacity of the intersection in the peak hour.

Rancang Bangun Sistem Informasi Situs Bersejarah Berbasis Web

Historic sites are a form of heritage and cultural heritage of ancestors that have value for the life of the nation today and in the future. Magelang City is one city that has many historic sites that have not been optimally managed. The k-medoid algorithm is a method used in creating groups based on objects as representatives (medoid) which function as cluster centers for each cluster. In this research, web-based historic site information system design in Magelang City is carried out. Through the information system that has been built to provide a friendly user information retrieval that has many advantages such as searching, grouping and information data accessed by visitors.