External Metrics Research Articles

Variation in Physical Performance of Futsal Players During Congested Fixtures.

To investigate the match-to-match variation of physical performance during official congested fixtures in elite futsal players. Physical performance was measured by external and internal load metrics in 12 elite male futsal players. Two periods with 3 matches within 4 days were analyzed. The variation in physical performance of the players during matches was analyzed using the latent growth curve modeling that estimated interindividual and intraindividual growth paths. Playing time had a significant effect on physical performance growth with significant paths of interindividual and intraindividual variability. Players who competed for more time revealed lower initial levels (ie,first match) of total distance covered (β = -0.62), high-speed running (β = -0.18), accelerations (β = -0.31), decelerations (β = -0.44), and session rate of perceived exertion (β = 0.81) than players who competed for less time (P < .05). In addition, players who competed for more time revealed higher increases in total distance covered (β = 0.47), high-speed running (β = 0.16), and session rate of perceived exertion (β = 0.66) and lower increases in accelerations (β = -0.21) and decelerations (β = -0.58) than players who competed for less time from the first to the third match (P < .05). Congested fixtures did not affect physical performance in elite futsal players. Playing time showed to be a key performance factor. There was a considerable heterogeneity in the responsiveness to physical performance over congested fixtures, suggesting an analysis of individual variability to evaluate real changes in match performance, training intensity, and workload.

Comparison of In-game External Load Metrics Among Positions and Between Halves for Division I Collegiate Women’s Lacrosse Athletes

External load has become a common metric for coaches to track the activity profiles of athletes during training and competition. The advent of wearable technology has made external load monitoring accessible for more coaches. The purpose of this study was to compare positional (attack, midfield, and defense) and game (first half to second) external loads. An NCAA Division I women’s lacrosse team was recruited to wear triaxal accelerometers and GPS units during five non-conference games during the 2020 regular season. The external load metrics evaluated for this study included total distance, sprint distance (&gt; 19 km∙hr-1), number of power plays (&gt; 3 m·s-2), top speed, and PlayerLoad. Significance was set at p &lt; 0.05. No significant differences among positions were observed for full game measures (p &gt; 0.05). A significant main effect for time was observed for sprint distance (midfield; p &lt; 0.001 ) and power plays (midfield; p &lt; 0.001 and defense; p = 0.004). While no significant differences occurred for activity profiles among positions, high-intensity efforts (sprint distance and power plays) were significantly less in the second half, likely due to fatigue. Coaches and sports scientists can use this information to manage in-game fatigue through tactics such as strategic substitutions and time-outs, thus preserving the intensity of the activity profiles late in the game.

Wearable Inertial Measurement Unit to Measure External Load: A Full-Season Study in Professional Soccer Players

The aim of this study was to describe weekly acute workload (wAW), chronic workload (wCW), acute: chronic workload ratio (wACWR), training monotony (wTM), and training strain (wTS) variations over a full season across playing positions. Twenty-one professional soccer players were daily monitored during 48 consecutive weeks. Total distance, sprint total distance (STD), high-speed running distance (HSRd), maximum speed, number of the repeated sprints, and body load (BL) were obtained during training and matches using a Wearable Inertial Measurement Unit. The wAW was determined for each external load measure. The wCW, wACWR, and wTM were calculated based on BL metric. Higher values of weekly STD were observed in lateral defenders/wingers (LDW) compared to central defenders/forwards (CDF) (p = 0.009; ES = Large) and midfielders (MDF) (p = 0.034; ES = Large). Additionally, weekly HSRd was higher in LDW vs. CDF (p = 0.016; ES = Large) and MDF (p = 0.011; ES = Large). The CDF presented a lower weekly number of repeated sprints than LDW (p = 0.021; ES = Large). In conclusion, weekly external load metrics were position-dependent over the season. Moreover, LDW a presented greater weekly STD, HSRd, and number of repeated sprints compared to other positions.

The Influence of Small-Sided Football Games with Numerical Variability in External Training Load

Small-sided games (SSGs) are a motivational strategy and effective training method to develop skills and physical fitness at optimal intensity. This study investigates the influence of variability in the number of players on the physiological response using SSGs. The sample was composed of 10 field players with an average age of 15.9 ± 0.50 years old. Three sessions were held on 3 different days to collect information. The 10 min SSG exercise with temporal numerical variability was repeated twice on each day. The numerical ratio of players involved changed every 2 min without disrupting the practice. In a game space with 35 × 25 m, the exercise always started in a situation of 3 vs. 3. External training load metrics (i.e., physical activity counts, activity intensity, and energy expenditure) were assessed using Actigraph WGT3X accelerometers. The 3 vs. 3 format game had higher physical activity counts (i.e., higher vector magnitude) and requested more time spent in very vigorous physical activity intensity. Consequently, the 3 vs. 3 format in young football players had higher energy expenditure (i.e., a higher METs) than that in the other studied game formats. This study confirms the influence of SSGs with numerical variability in external training load.

Athletic Monitoring in Basketball: A Qualitative Exploratory Approach

The competitiveness of sport continues to increase with the corollary increase in sport technology. Sport technology is an umbrella that encompasses biometrics, wearable technology, and data analytics. Central to sport technology is a close relative, sport analytics. If sport technology is the general term for measuring what the athlete is experiencing from a performance standpoint, then sport analytics is the process of cleaning, processing, and manipulating these variables into meaningful output. The number of variables that can be collected are extensive but can be generally categorized into internal and external metrics. Internal metrics seek to measure how the athlete responds to training load (HR, HRV, Hormones, Sleep patterns). External metrics seek to measure how much work the athlete is performing (Live player tracking, Countermovement Jump, Isometric Mid-Thigh Pull). Internal and external metrics are continually evolving in the manner they are collected which depends on the time, resources, and staff allocation of an organization. This is where the collaboration between sport scientists and sport performance staff can come together to overlay variables and find relationships. There is no correct way to analyze data because each organization will have their own goals. For example, data could be used for injury mitigation, readiness assessment, or simply to establish baseline values. Some common statistical techniques that can be used include correlation, regression, and principal component analysis (PCA). However, performing these calculations may be complex and some organizations may lack a practitioner which can limit their ability to use data. In seeing this gap, the authors propose a qualitative approach to workload monitoring made for basketball but can be applied to most sports. The purpose of this paper is to empower sports performance staffs to learn and implement the basics of qualitative athletic monitoring approach.

An evaluation of internal and external workload metrics in games in women’s collegiate lacrosse

An evaluation of internal and external workload metrics in games in women’s collegiate lacrosse

Assessment of Relationships Between External Load Metrics and Game Performance in Women’s Lacrosse

Assessment of Relationships Between External Load Metrics and Game Performance in Women’s Lacrosse

Practice versus Game External Load Measures in Starters and Non-Starters of a Men’s Collegiate Soccer Team

Monitoring external loads may minimize injury risk and improve physical performance. The purpose was to describe the external loads of a men’s collegiate soccer team during practice and games at the start of in-season play. In the first 2 weeks of the competitive season, National Collegiate Athletic Association Division I soccer athletes (n=19; mean±SD, age: 20.3±0.9 yr; body mass: 77.9±6.8kg; body height: 178.87±7.18cm; body fat: 10.0±5.0%; VO2max: 65.39±7.61mL/kg/min) wore a global positional system device (GPS/GNSS) during practices (n=8) and games (n=3). Starters were classified as players who maintained a minimum playing time of 45 minutes per game (n=10); other players were considered non-starters (n=9). External load metrics collected were: total distance (TD), player load (PL), high-speed distance (HSD, &gt;13 mph (5.8 m/s)), high inertial movement analysis (IMA, &gt;3.5m/s2), and repeated high intensity efforts (RHIE). Multivariate and repeated measures analyses of variance assessed differences in external load measures for practices and games in starters and non-starters. Relative to game loads, practices were quantified as high (&gt;1 SD above the mean), medium (1 SD below the mean), low (2 SD below the mean) and very low (3 SD below the mean). For starters and non-starters, TD, PL, HSD, IMA, and RHIEs were lower in practices compared to games (p&lt;0.001). No practices were classified as high or medium for any external load measure, with the majority of practices (75-100%) being classified as very low. Therefore, practice did not simulate game volumes or intensities. An individualized approach to monitoring is recommended to ensure starters receive adequate recovery and non-starters receive exposure to game-load physical stress.

Validation of population pharmacokinetic models: a comparison of internal and external validation approaches for hydrochlorothiazide

1. Model evaluation is an important issue in population analyses. Our aim was to perform and illustrate metrics and techniques for internal and external evaluation with an application to population pharmacokinetics of hydrochlorothiazide (HCTZ). 2. A nonlinear mixed effects model was used to study the pharmacokinetics of HCTZ. In addition, different types of internal assessment tools and external metrics were used for model evaluation. External evaluation was performed using an alternative dataset that included data from an independent group of subjects. For comparison, a previously published population pharmacokinetic model for HCTZ was applied to the same data. 3. A two-compartment model with first-order oral absorption using a constant time delay between administration and absorption and first-order elimination best described HCTZ pharmacokinetics. Age had a statistically significant effect on HCTZ clearance. The final model performed adequately in the internal and external assessment tests. The final model showed better predictive performance than the other previously published HCTZ model. 4. Finally, a robust population pharmacokinetic model for HCTZ in adults was constructed and validated internally and externally. Incorporating analytical assessment of nonlinear pharmacokinetics into the modelling may be a promising approach to improve the predictive power of the model.

Grey wolf optimization algorithm for hierarchical document clustering

In data mining, the application of grey wolf optimization (GWO) algorithm has been used in several learning approaches because of its simplicity in adapting to different application domains. Most recent works that concern unsupervised learning have focused on text clustering, where the GWO algorithm shows promising results. Although GWO has great potential in performing text clustering, it has limitations in dealing with outlier documents and noise data. This research introduces medoid GWO (M-GWO) algorithm, which incorporates a medoid recalculation process to share the information of medoids among the three best wolves and the rest of the population. This improvement aims to find the best set of medoids during the algorithm run and increases the exploitation search to find more local regions in the search space. Experimental results obtained from using well-known algorithms, such as genetic, firefly, GWO, and k-means algorithms, in four benchmarks. The results of external evaluation metrics, such as rand, purity, F-measure, and entropy, indicates that the proposed M-GWO algorithm achieves better document clustering than all other algorithms (i.e., 75% better when using Rand metric, 50% better than all algorithm based on purity metric, 75% better than all algorithms using F-measure metric, and 100% based on entropy metric).

Discerning the Management-Relevant Ecology and Distribution of Sea Pens (Cnidaria: Pennatulacea) in Norway and Beyond

Sea pens are considered to be of conservation relevance according to multiple international legislations and agreements. Consequently, any information about their ecology and distribution should be of use to management decision makers. This study aims to provide such information about six taxa of sea pen in Norwegian waters [Funiculina quadrangularis (Pallas, 1766), Halipteris spp., Kophobelemnon stelliferum (Müller, 1776), Pennatulidae spp., Umbellula spp., and Virgulariidae spp.]. Data exploration techniques and ensembled species distribution modelling (SDM) are applied to video observations obtained by the MAREANO project between 2006 and 2020. Norway-based ecological profiles and predicted distributions are provided and discussed. External validations and uncertainty metrics highlight model weaknesses (overfitting, limited training/external observations) and consistencies relevant to marine management. Comparison to international literature further identifies globally relevant findings: (a) disparities in the environmental profile of F. quadrangularis suggest differing “realised niches” in different locations, potentially highlighting this taxon as particularly vulnerable to impact, (b) none of the six sea pen taxa were found to consistently co-occur, instead partially overlapping environmental profiles suggests that grouping taxa as “sea pens and burrowing megafauna” should be done with caution post-analyses only, (c) higher taxonomic level groupings, while sometimes necessary due to identification issues, result in poorer quality predictive models and may mask the occurrence of rarer species. Community-based groupings are therefore preferable due to confirmed shared ecological niches while greater value should be placed on accurate species ID to support management efforts.

Chemometric modeling of plant protection products (PPPs) for the prediction of acute contact toxicity against honey bees (A. mellifera): A 2D-QSAR approach

Honey bees (Apis mellifera) are vital for economic, viable agriculture and for food safety. Although Plant Protection Products (PPPs) are of undeniable importance in the global agricultural system, these have become potential threats for non-target organisms like pollinators (e.g., honey bees etc.), resulting in the disruption of the ecological balance. In the current work, we have used the 113 PPP analogs to develop a 2D-QSAR model and explored the structural features modulating the toxic effects on honey bees, following the Organization for Economic Co-operation and Development (OECD) guidelines. The extensive validation of the developed model has been performed using internal and external validation metrics to make sure that the model is statistically sound and interpretable enough to be acceptable. The obtained results (R2 = 0.666, Q2 = 0.594, Q2F1 = 0.647 and Q2F2 = 0.646) determine the predictability and reliability of the developed model. This model should be useful for the predictions (acute contact toxicity (LD50)) of the new and untested compounds located inside the applicability domain of the developed model. Moreover, we have performed the in-silico prediction of toxicity against honey bees of a total of 709 compounds obtained from the pesticide properties database (PPDB) using the developed model.

J-EDA: A workbench for tuning similarity and diversity search parameters in content-based image retrieval

Similarity searches can be modeled by means of distances following the Metric Spaces Theory and constitute a fast and explainable query mechanism behind content-based image retrieval (CBIR) tasks. However, classical distance-based queries, e.g., Range and k-Nearest Neighbors, may be unsuitable for exploring large datasets because the retrieved elements are often similar among themselves. Although similarity searching is enriched with the imposition of rules to foster result diversification, the fine-tuning of the diversity query is still an open issue, which is is usually carried out with and a non-optimal expensive computational inspection. This paper introduces J-EDA, a practical workbench implemented in Java that supports the tuning of similarity and diversity search parameters by enabling the automatic and parallel exploration of multiple search settings regarding a user-posed content-based image retrieval task. J-EDA implements a wide variety of classical and diversity-driven search queries, as well as many CBIR settings such as feature extractors for images, distance functions, and relevance feedback techniques. Accordingly, users can define multiple query settings and inspect their performances for spotting the most suitable parameterization for a content-based image retrieval problem at hand. The workbench reports the experimental performances with several internal and external evaluation metrics such as P × R and Mean Average Precision (mAP), which are calculated towards either incremental or batch procedures performed with or without human interaction.

Arsenic exposure analysis for children living in central China: From ingestion exposure to biomarkers

Emerging evidence indicates that chronic low-dose arsenic (As) exposure can pose adverse health effects to children. This study aimed to systematically study the exposure risk induced by As ingestion in children living in Hubei Province, central China. The feasibility of first morning spot urine instead of 24-h urine as an environmental exposure biomarker was also explored. A total of 120 children aged 2–17 years were recruited from an urban area for the collection of biomarker samples (first morning and 24-h urine samples), environmental exposure samples (duplicate diets, drinking water, and soil), and related child-specific exposure factors. The external exposure risk, internal exposure level, and source of exposure to As in children were analyzed. The results indicated that As concentration in duplicated diets, water, and soil were 29.2 μg kg−1, 1.3 μg L−1, and 9.3 mg kg−1, respectively; these were all below the corresponding maximum allowable levels in China (the threshold value of As in most food, drinking water and soil are 0.5 mg⸱kg−1, 0.01 mg L−1, and 20 mg⸱kg−1, respectively). Dietary intake was the predominant exposure route, accounting for 90% of the total daily dose. The combined oral non-carcinogenic and carcinogenic risks all exceeded the corresponding maximum acceptable risk level. Therefore, As bioavailability should be investigated and used in health risk assessment. Multiple linear regression analysis indicated that urinary As was positively associated with dietary As, with a one-unit increase in daily As intake from the diet associating with 4.82 and 5.21 μg g−1 increases in 24-h urine and first morning urine, respectively. Furthermore, significant correlations with 24-h urine and external exposure metrics suggested that creatine-adjusted As concentrations in first morning urine could be an appropriate substitute of 24-h urine as exposure biomarkers.

QSTR and interspecies-QSTR modelling for aquatic toxicity data gap filling of cationic polymers

Polymers are extensively used in several fields representing a growing multi-billion dollar industry covering several thousands of materials and billions of kilos used globally every year. The widespread use of cationic polymers may significantly discharge such compounds into the aquatic environment, which may cause potential toxic effects on aquatic organisms. The amount of publicly available, high-quality environmental toxicity data for industrial polymers such as cationic polyquaterniums is low. We have developed here individual quantitative structure–toxicity relationship (QSTR) models for toxicity prediction against fish and algae. These models against fish and algae showed optimistic statistical quality in terms of several internal and external quality and validation metrics such as determination coefficient R2 (0.703 and 0.676), cross-validated leave-one-out Q2 (0.638 and 0.516) and predictive R2pred or Q2ext (0.776 and 0.703) for fish (Ntrain = 72, Ntest = 23) and algae (Ntrain = 40, Ntest = 14) toxicity datasets, respectively. The study revealed that higher charge density increases the toxicity against both the response endpoints. However, a higher percentage of oligomers with a molecular mass of lower than 1000 Daltons results in a decreased toxicity towards both the studied endpoints. Similarly, primary amines in the molecular building block result in a reduction in the toxicity against the algal species. However, acceptable individual QSTR models against D. magna could not be generated with the limited feature information obtained from the United States Environmental Protection Agency and additional data provided by Environmental Climate Change Canada (ECCC). Therefore, we have also proposed interspecies quantitative structure–toxicity relationship (i-QSTR) models among three species (D. magna, fish and algae species) to bridge the toxicity data gap for cationic polymers. The mechanistic interpretation of i-QSTR models revealed several important characteristic features of polymers along with the experimental response of one species which are also helpful for toxicity prediction of other species, ultimately helping to reduce the experimental testing for toxicity prediction. Finally, the proposed QSTR and i-QSTR models can be helpful to compute the toxicity of polymers in the early stages of screening for regulatory purposes and data gap filling for new or untested polymers falling within the applicability domain of the models.

External and Internal Load Measures During Preseason Training in Men Collegiate Soccer Athletes.

Fields, JB, Merigan, JM, Gallo, S, White, JB, and Jones, MT. External and internal load measures during preseason training in men collegiate soccer athletes. J Strength Cond Res 35(9): 2572-2578, 2021-Collegiate athletes are exposed to high volume loads during preseason training. Monitoring training load can inform training and recovery periods. Therefore, the purpose was to examine changes in and bidirectional relationship between external and internal load metrics in men collegiate soccer athletes (n = 20; age, 20 ± 1 year). Internal load measures of heart rate variability (HRV), salivary testosterone (T) and cortisol (C), and self-assessment wellness and ratings of perceived exertion scales were collected daily. External load measures of total distance, player load, high-speed distance, high inertial movement analysis, and repeated high-intensity efforts were collected in each training session using global positioning system/global navigation satellite system technology. A 1-way analysis of variance determined weekly changes in external load, physiological, hormonal, and subjective self-assessment measures of internal load. Bidirectional prediction of external load markers and self-assessment measures on physiological and hormonal markers of internal load were assessed by hierarchical linear regression models (p < 0.05). External load measures, C, energy, sleep, and rate of perceived exertion (RPE) decreased (p < 0.01), whereas T, T:C ratio, anger, depression, and vigor increased (p < 0.01) from week 1 to week 2. Morning C positively predicted afternoon external load and post-training RPE (p < 0.05); T:C ratio negatively predicted afternoon external load and post-training RPE (p < 0.05); and morning HRV negatively predicted post-training RPE (p = 0.031). Despite reduced hormonal stress and external load across weeks, negative perceptions of fatigue increased, suggesting fatigue patterns may have a delayed response. Load may have a more belated, chronic effect on perceptions of fatigue, whereas hormonal changes may be more immediate and sensitive to change. Practitioners may wish to use a variety of external and internal load measures to understand athletes' stress responses to training.

Development of QSAR models for evaluating pesticide toxicity against Skeletonema costatum

Nowadays, the emergence of pesticides and its application in agriculture greatly improved the crop quality and food production. However, the resulted ecological problem caused by the widespread pesticide residues attracted more and more attention since the pesticides were harmful to most living organisms. Regulatory agencies such as Environmental Protection Agency (EPA) and European Chemicals Agency (ECHA) stipulated that a comprehensive pesticides risk assessment was essential and also underscored the application of computation method in evaluating pesticides. The present study aimed to use the Quantitative Structure-Activity Relationship (QSAR) method to establish models for quantitatively and qualitatively predicting the toxicity of pesticide against Skeletonema costatum. The regression model was developed using the Genetic Algorithm plus Multiple Linear Regression method and the classification model was established based on the Random Forest algorithm, respectively. Various internal and external validation metrics suggested that the obtained regression model was of good fitness (R2=0.722), robustness (QLOO2=0.653) and external predictive ability (QFn2:0.719–0.776, CCC = 0.878). The classification could correctly predict 79.4% of pesticides in the training set and 69.7% in the validation set. The relatively high sensitivity value of the classification model indicated its good performance in identifying high-toxic pesticides. It could be concluded from the selected modelling descriptors that molecular weight and polarizability impacted the toxicity the most. The atom-type E-state descriptors generally contributed negatively to the pesticide toxicity which verified the negative influence of molecular hydrophilicity. Moreover, the lipophilic, carbon-type, charge related descriptors demonstrated the important influence of lipophilicity and polarity on pesticide toxicity. The models presented in this work could be used to pre-evaluate the toxicity of pesticides within the applicability domain, thus focusing resources on the high-toxic pesticides and assessing the environmental risk of pesticides quickly and economically.

The Impact of Network Topology on Air Transportation Robustness to Pandemics

The COVID-19 pandemic has brought a brand-new challenge to the design and management of the air transportation network (ATN). Many countries have partially or entirely shut down their international flights during this period to contain the spread of COVID-19. However, the changes in the international connectivity of different countries during this period varied greatly, as the ATNs of different countries have different structures. Accordingly, this paper studies the impact of network topology on the international robustness of a country's ATN to pandemic-induced disruptions. Therein, a Weighted International Efficiency metric is proposed to evaluate a country's connectivity to other countries, followed by deriving an International Robustness metric that measures the ability of a country's ATN to maintain international connections during pandemics. Further, three internal and external topological metrics are defined to characterize the structure of an ATN. The results of correlation and regression analysis show that the proportion of international airports inside a country is the most important factor in determining its international robustness to pandemics, followed by the density of the domestic connections. Surprisingly, the number of directly connected foreign cities has little effect on international robustness. The analysis demonstrates the importance of studying ATN topology during emergencies.

Integrating External and Internal Load for Monitoring Fitness and Fatigue Status in Standard Microcycles in Elite Rink Hockey.

The aims of this study were 3-fold: firstly, to present an integrative approach to external and internal load dynamics for monitoring fitness and fatigue status of specific in-court rink hockey training sessions in a standard microcycle; secondly, to assess the differences between training sessions and matches; the third and final aim was to assess the association between external and internal load metrics. The external load, using a local positioning system, and internal load, using the declared rate of perceived exertion, were measured during 23 in-season microcycles for nine top-level players. Training load data were analysed with regard to the number of days before or after a match [match day (MD) minus or plus]. In relation to the first aim, internal and external load metrics merged into a single integrated system using pooled data z-scores provided an invisible monitoring tool that places the players in the fitness-fatigue continuum throughout the different microcycle sessions. In this regard, MD-4 and MD-1 sessions tend to place, with a low dispersion, the players in a “low external and internal load” zone. On the contrary, in MD-3 and MD-2 sessions, as well as in MD, in which higher loads were recorded, most of the players were within a “high external and internal load” zone with a tendency towards dispersion towards the fitness or fatigue zones. Finally, and with regard to the second and third aims, an inverted “U-shape” load dynamic related to the specific goals of each training session was the main finding in terms of comparison between MD; a load peak between MD-3 and MD-2 sessions and a significant decrease in all the load variables in MD-1 sessions were found; and high-to-low correlations were found between external and internal load metrics. This study presents an integrative approach to the external and internal load of players for monitoring fitness and fatigue status during a standard microcycle in rink hockey that might provide team sport staff members with a deeper understanding of load distribution in the microcycle in relation to the match.

Improved Self-Adaptive ACS Algorithm to Determine the Optimal Number of Clusters

A fundamental problem in data clustering is how to determine the correct number of clusters. The k-adaptive medoid set ant colony optimization (ACO) clustering (METACOC-K) algorithm is superior in solving clustering problems. However, METACOC-K does not guarantee in finding the best number of clusters. It assumed the number of clusters based on an adaptive parameter strategy that lacks feedback learning. This has restrained the algorithm in producing compact clusters and the optimal number of clusters. In this paper, a self-adaptive ACO clustering (S-ACOC) algorithm is proposed to produce the optimal number of clusters by incorporating a self-adaptive parameter strategy. The S-ACOC algorithm is a centroid-based algorithm that automatically adjusts the number of clusters during the algorithm run. The selection of the number of clusters is based on a construction graph that reflects the influence of a pheromone in algorithm learning. Experiments were conducted on real-world datasets to evaluate the performance of the proposed algorithm. The external evaluation metrics (purity, F-measure, and entropy) were used to compare the results of the proposed algorithm with other swarm clustering algorithms, including a genetic algorithm (GA), particle swarm optimization (PSO), and METACOC-K. Results showed that S-ACOC provides higher purity (50%) and lower entropy (40%) than GA, PSO, and METACOC-K. Experiments were also performed on several predefined clusters, and results demonstrate that the S-ACOC algorithm is superior to GA, PSO, and METACOC-K. Based on the superior performance, S-ACOC can be used to solve clustering problems in various application domains.