Firefly Algorithm Research Articles

Hierarchical learning multi-objective firefly algorithm for high-dimensional feature selection

Feature selection is a crucial data preprocessing technique extensively employed in machine learning and image processing. However, feature selection encounters significant challenges when addressing high-dimensional data due to the huge and discrete decision space. This paper proposes a hierarchical learning multi-objective firefly algorithm (HMOFA) for solving the feature selection task in high-dimensional data. The main contributions are as follows: 1) Features are clustered based on the evaluation of multiple metrics, which are used to initialize the population and improve the quality of the initial population; 2) A hierarchy-guided learning model is proposed, where individuals move toward superior solutions while moving away from inferior solutions, avoiding the oscillation phenomenon that occurs under the full attraction model, and reducing the likelihood of the population being trapped in a local optimum; 3) Use duplicate solution modification mechanism to reduce the number of duplicate individuals in the population. The proposed method is compared with 8 competitive feature selection methods using 15 datasets, and the results demonstrate that HMOFA can achieve higher classification accuracy while selecting fewer features.

Optimized neural network-based model to predict the shear strength of trapezoidal-corrugated steel webs

Beam-like members use corrugated webs to increase their shear strength, stability, and efficiency. The corrugation positively affects the members' structural characteristics, especially those governed by the web parameters, such as the shear strength, while reducing the total weight. Existing code and analytical models for predicting the shear strength of trapezoidal corrugated steel webs (TCSWs) are summarized. This paper presents an optimized Artificial Neural Network (ANN)-based model to estimate the shear strength of steel girders with a TCSW subjected to a concentrated force. A database of 206 experimental results from the literature is used to feed the ANNs. Six geometrical and material parameters were identified as input variables, and the experimental shear strength at failure was considered the output variable. Four hyperparameter optimization techniques are applied to refine the ANN models: Bayesian Optimization (BO), Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS), Firefly Algorithm (FA), and African Buffalo Optimization (ABO). The performance metrics indicate that the ABO-ANN model is the most effective among these. The predictions of the developed ML model were also compared with those of existing code and analytical models. The comparisons illustrated that the ANN-based model outperforms the other existing models. The sensitivity analysis using the proposed ANN-based model captured the relationships and interactions among the geometric and material parameters and their impact on shear strength. One main finding is that the corrugation angle in the 35–45° range maximized the TCSW shear strength.

Optimal planning and partitioning of multiple distribution Micro-Grids based on reliability evaluation.

Power system researcher have turned to Micro-Grids (MG) for higher reliability, greater flexibility, lower operating costs and losses, and lower CO2 emissions at the distribution system. This paper presents a single-level stochastic optimization framework for planning and partitioning of a distribution system including Multiple Micro-Grids (MMGs). The main objective is to minimize the total cost of the system including investment, operation, total losses and reliability costs of the distribution network. The proposed model takes into account the viewpoints of MG owners and distribution system operators, simultaneously. The voltage stability index is introduced to identify the optimal site of MG investment. To deal with uncertainties caused by renewable generations, the Firefly Algorithm (FA) and probability-tree method is used to create various operation scenarios of Photo-Voltaic (PVs) and Wind Turbines (WTs). This model is solved through the genetic algorithm in MATLAB, and to evaluate its effectiveness, numerical studies have been carried out on the experimental IEEE distribution network with four specified locations for investing MGs and seven Tie Switches (TS) for network partitioning. Simulation results reveal that optimal locations for MG investment are determined in such a way all MGs connect to the buses near the beginning of the feeder and as a result, load point reliability is improved, total active power losses are reduced, and the energy program becomes more optimized.

Optimization design of hydro turbine support structure based on GA-FA-BP method

The rational design of the turbine support structure ensures the safe operation of the system, achieves significant economic benefits and improves the efficiency of the turbine. This paper presents a novel turbine support structure capable of vertical movement along the guide column. In addition, the GA-FA-BP method has been developed for the optimization design of the support structure, which combines genetic algorithm (GA), firefly algorithm (FA) and back-propagation neural network (BP). The use of quadratic response surface (RS) method along with NSGA-II for the result validation of the GA-FA-BP model ensures the robustness and reliability of the optimization results. A finite element model is set up to analyse the mechanical behaviour of the support structure under 10 different combined load conditions. The most critical load conditions in the static mechanical models are used to generate training data, and the mass, deformation and equivalent stress of the optimized and unoptimized support structures are analyzed and compared. The results show that the optimized support structure can maintain deformation and equivalent stress within a weight reduction of 21.83% compared to the original design. Compared to a single optimization model, the proposed GA-FA-BP model shows higher accuracy with a correlation coefficient up to 0.9989.

Sediment load forecasting from a biomimetic optimization perspective: Firefly and Artificial Bee Colony algorithms empowered neural network modeling in Çoruh River

The service life of downstream dams, river hydraulics, waterworks construction, and reservoir management is significantly affected by the amount of sediment load (SL). This study combined models such as the artificial neural network (ANN) algorithm with the Firefly algorithm (FA) and Artificial Bee Colony (ABC) optimization techniques for the estimation of monthly SL values in the Çoruh River in Northeastern Turkey. The estimation of SL values was achieved using inputs of previous SL and streamflow values provided to the models. Various statistical metrics were used to evaluate the accuracy of the established hybrid and stand-alone models. The hybrid model is a novel approach for estimating sediment load based on various input variables. The results of the analysis determined that the ABC-ANN hybrid approach outperformed others in SL estimation. In this study, two combinations, M1 and M2, with different input variables, were used to assess the model's accuracy, and the best-performing model for monthly SL estimation was identified. Two scenarios, Q(t) and Q(t − 1), were coupled with the ABC-ANN algorithm, resulting in a highly effective hybrid approach with the best accuracy results (R2 = 0.90, RMSE = 1406.730, MAE = 769.545, MAPE = 5.861, MBE = − 251.090, Bias Factor = − 4.457, and KGE = 0.737) compared to other models. Furthermore, the utilization of FA and ABC optimization techniques facilitated the optimization of the ANN model parameters. The significant results demonstrated that the optimization and hybrid techniques provided the most effective outcomes in forecasting SL for both combination scenarios. As a result, the prediction outputs achieved higher accuracy than those of a stand-alone ANN model. The findings of this study can provide essential resources to various managers and policymakers for the management of water resources.

Retraction Note: Optimization using the firefly algorithm of ensemble neural networks with type-2 fuzzy integration for COVID-19 time series prediction.

[This retracts the article DOI: 10.1007/s00500-020-05549-5.].

Development and optimization of machine learning models for estimation of mechanical properties of linear low-density polyethylene

A hybrid methodology was developed and implemented for estimation of polymeric mechanical properties in rotational moulding process. The considered polymer in this study is linear low-density polyethylene, known as LLDPE, which has extensive application in plastic industry. The mechanical properties of the polymer were assessed and correlated to the oven residence time to build the predictive model of moulding process. A tiny dataset containing only 25 data rows via a number of machine learning models were assessed. Oven residence time is the only input, while the LLDPE's properties including tensile strength, impact strength, and flexure strength are the outputs considered in the machine learning models. We used tree-based ensemble methods for modeling in this work and they are tuned using FA (Firefly Algorithm) optimizer to find optimal hyper-parameters of them. Finally, the optimal models had shown a great performance to predict the output accurately. For tensile strength, the best model (FA-ET) has an R2 value of 0.9994, this score is 0.9995 for impact strength and 0.9968 for flexure strength. The tree-based models tuned in this study revealed to be robust in estimation of polymeric properties and can be used to obtain the products with the best quality.

Detection of COVID-19: A Metaheuristic-Optimized Maximally Stable Extremal Regions Approach

The challenges associated with conventional methods of COVID-19 detection have prompted the exploration of alternative approaches, including the analysis of lung X-ray images. This paper introduces a novel algorithm designed to identify abnormalities in X-ray images indicative of COVID-19 by combining the maximally stable extremal regions (MSER) method with metaheuristic algorithms. The MSER method is efficient and effective under various adverse conditions, utilizing symmetry as a key property to detect regions despite changes in scaling or lighting. However, calibrating the MSER method is challenging. Our approach transforms this calibration into an optimization task, employing metaheuristic algorithms such as Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), Firefly (FF), and Genetic Algorithms (GA) to find the optimal parameters for MSER. By automating the calibration process through metaheuristic optimization, we overcome the primary disadvantage of the MSER method. This innovative combination enables precise detection of abnormal regions characteristic of COVID-19 without the need for extensive datasets of labeled training images, unlike deep learning methods. Our methodology was rigorously tested across multiple databases, and the detection quality was evaluated using various indices. The experimental results demonstrate the robust capability of our algorithm to support healthcare professionals in accurately detecting COVID-19, highlighting its significant potential and effectiveness as a practical and efficient alternative for medical diagnostics and precise image analysis.

Quantum-inspired firefly algorithm with ant miner plus for fake news detection

Nowadays, technology has shifted the way individuals access news from conventional media sources to social media platforms. The active engagement of people with social media platforms leads them to consume news without confirming its source or legitimacy. This facilitated the dissemination of more manipulated and false information in the form of rumors and fake news. Fake news can affect public opinion and create chaos and panic among the population. Thus, it is essential to employ an advanced methodology to identify fake news with high precision. This research work has proposed the concept of the quantum-inspired firefly algorithm with the ant miner plus algorithm (QFAMP) for more effective fake news detection. The proposed QFAMP algorithm utilizes the attributes of quantum computing (QC), the firefly algorithm (FA), and the ant miner plus algorithm (AMP). Here, the QFA approach ensures the effective exploitation of the firefly agents until the agents are able to search for the brighter firefly. Further, the AMP algorithm utilizes the best ants with higher pheromone concentrations for global exploration, which also avoids the premature convergence of the QFA agents. In addition, the AMP algorithm serves as an efficient data mining variant that is effective for the classification of fake news. The efficacy of the proposed QFAMP algorithm is evaluated for the dataset of FakeNewsNet, which is composed of two sub-categories: BuzzFeed and PolitiFact. The experimental evaluations indicate the effective performance of the proposed algorithm compared to the other techniques.

Energy Efficient Data Aggregation in Wireless Sensor Networks Using Meta Heuristic Based Feed Forward Back Propagation Neural Network Approach

Sensor nodes are low-cost, low-power, tiny devices that make up the majority of WSNs, or distributed, self-organizing systems. These sensor nodes are able to exchange, perceive, and interpret data. The sensor nodes are equipped with a wide variety of sensors, such as chemical, touch, motion, temperature, and weather sensors. Because of its adaptability, sensors are used in a variety of applications such as automation, tracking, monitoring, and surveillance. Despite the enormous number of sensor applications, WSNs continue to suffer from common challenges like as low memory, slow processing speed, and short network lifetime. The feed forward back propagation neural network mode (FFBPNN) based on meta heuristics aims to create many paths for effective data aggregation in wireless sensor networks. This model handled the process of identifying and selecting the optimum route path. The distributed sensor nodes are utilized to create the various route paths. In this research paper, data aggregation is done using meta-heuristic firefly algorithm that helped in identifying an optimal route from among the found routes. After selecting the operative ideal route choice, the data aggregation procedure practices a rank-based approach to accomplish lower latency and a better packet delivery ratio(PDR). In addition to throughput, simulation was done to improve and measure performance in terms of packet delivery ratio, energy consumption, and end-to-end latency.

Efficient Heart Disease Classification Through Stacked Ensemble with Optimized Firefly Feature Selection

In the current century, heart-related sickness is one of the important causes of death for all humans. An estimated 17.5 million deaths occur due to heart disease worldwide. It is observed that more than 75% of peoples with average income level mostly suffer from heart diseases and its complications. So, there is need for predicting heart infection and its related complications. Data mining is the method of converting raw data into useful information. These tools allow given data to predict future trends. Data mining concepts were mainly adapted in heart disease data sets to interpret the intricate inferences out of it. In the modern world, many research are carried in health care engineering with the use of mining and prediction techniques. This investigation aims to identify significant features in heart disease dataset and to apply ensembling techniques for improving exactness of prediction. Prediction models are developed using different ensembling techniques like stacking and voting. For the experimental purpose, the Z-Alizadeh Sani dataset is used, which is available in the UCI machine learning data repository. Stacking and voting techniques are applied to the dataset. Stacking with substantial characteristics has the maximum accuracy of 86.79% in the Z-Alizadeh dataset. Test outcome proves that the prediction model implemented with the features selected using firefly algorithm and stacking-based classification model has the highest accuracy prediction than other technique. Furthermore, this study delineates a comparative analysis with prior works, showcasing the superior capabilities of the firefly algorithm in optimizing feature selection processes, which is crucial for advancing the accuracy of heart disease predictions.

Improving load frequency controller tuning with rat swarm optimization and porpoising feature detection for enhanced power system stability

Load frequency control (LFC) plays a critical role in ensuring the reliable and stable operation of power plants and maintaining a quality power supply to consumers. In control engineering, an oscillatory behavior exhibited by a system in response to control actions is referred to as “Porpoising”. This article focused on investigating the causes of the porpoising phenomenon in the context of LFC. This paper introduces a novel methodology for enhancing the performance of load frequency controllers in power systems by employing rat swarm optimization (RSO) for tuning and detecting the porpoising feature to ensure stability. The study focuses on a single-area thermal power generating station (TPGS) subjected to a 1% load demand change, employing MATLAB simulations for analysis. The proposed RSO-based PID controller is compared against traditional methods such as the firefly algorithm (FFA) and Ziegler-Nichols (ZN) technique. Results indicate that the RSO-based PID controller exhibits superior performance, achieving zero frequency error, reduced negative peak overshoot, and faster settling time compared to other methods. Furthermore, the paper investigates the porpoising phenomenon in PID controllers, analyzing the location of poles in the s-plane, damping ratio, and control actions. The RSO-based PID controller demonstrates enhanced stability and resistance to porpoising, making it a promising solution for power system control. Future research will focus on real-time implementation and broader applications across different control systems.

Optimal sizing of a fixed-tilt ground-mounted grid-connected photovoltaic system with bifacial modules using Harris Hawks Optimization

This paper presents an optimal design for ground-mounted grid-connected bifacial PV power plants using a Computational Intelligence (CI)- based Harris Hawks Optimization (HHO) algorithm. This HHO algorithm identifies the best configuration of components and installation parameters for the bifacial PV power plant, aiming to maximize the final yield, minimize the Levelized Cost of Electricity, and boost the Net Present Value. Four variables were optimized: the bifacial PV module model, inverter model, tilt angle, and module elevation. Furthermore, the paper introduces a Harris Hawks Optimization Sizing Algorithm (HHOSA) to address the sizing challenges. The presented HHOSA was purely developed in Matlab R2017b. The usage of PVsyst was only limited to the derivation of irradiation data at different tilt angle of PV array. These data were later used in HHOSA. To verify its effectiveness, HHOSA was benchmarked against other CI algorithms, including the Slime Mould Algorithm (SMA), Firefly Algorithm (FA), Manta Ray Foraging Optimization (MRFO), and Cuckoo Search Algorithm (COA). The evaluation considered the algorithm’s stability, local search capability, convergence rate, computation time, and required population size. Findings suggest that the HHOSA outperforms its peers, marking it as a potential leader for designing bifacial PV power plants. The results indicate that the HHOSA algorithm exhibits superior performance in these aspects, making it a promising approach for optimizing the design of bifacial PV power plants. Moreover, this study provides insights into the economic and technical viability of bifacial PV systems under various environmental and system conditions. A sensitivity analysis, focusing on the interplay of three decision variables − albedo values (25 %, 50 %, and 75 %), tilt angles (10°, 25°, and 35°), and module elevations (0.5 m, 1.5 m, and 2 m) − was conducted. It assessed their influence on final yield, additional bifacial PV module yield, Levelized Cost of Electricity, and the system’s Net Present Value. The results emphasize the importance of carefully considering the impacts of albedo, module elevation, and tilt angle on the financial performance of bifacial PV installations.

Cooperative path planning study of distributed multi-mobile robots based on optimised ACO algorithm

The rapid development of robotics technology has driven the growth of robot types and the development of related technologies. As an important aspect of robot research, path planning technology plays an irreplaceable role in practical production and application. Ant colony algorithm has a wide range of applications in robot path planning, but there is also a problem of performance overly relying on initial parameter selection. In order to solve this problem and improve the performance of mobile robot path planning, an improved ant colony algorithm based on firefly algorithm was studied and designed in a two-dimensional environment. In order to further explore the performance of ant colony algorithm in solving robot coordinated path planning problems, an improved ant colony algorithm based on heuristic function was also designed. In a three-dimensional environment, an improved ant colony algorithm based on the improved artificial potential field method was designed. The research results show that the maximum running time of the improved ant colony algorithm based on the firefly algorithm in different grid environments is 819.36 s, 847.01 s, and 811.54 s, respectively. The average running time of the improved ant colony algorithm based on heuristic function in different grid environments is 5.19 s, 5.97 s, and 9.09 s, with average path lengths of 29.90 cm, 31.08 cm, and 37.01 cm, and path length variances of 0.35, 0.87, and 2.21, respectively. The ant colony algorithm based on the improved artificial potential field method has a running time of 1.930 s, 3.182 s, and 4.662 s in different grid environments, and a path length of 29.275 cm, 49.447 cm, and 67.057 cm, respectively. The ant colony algorithm for research and design optimization has good performance. The contribution of the research lies in the design of three path planning methods for mobile robots, including two-dimensional path planning and three-dimensional path planning, which improves the time of path planning and shortens the average path length. The novelty of the research is reflected in the design of a path planning method for mobile robots in two-dimensional and three-dimensional environments, which improves the ant colony algorithm through firefly algorithm and heuristic function, and combines the ant colony algorithm with the improved artificial potential field method. The method designed by the research institute can provide technical support for path planning of mobile robots.

Comparative analysis of various machine learning algorithms to predict strength properties of sustainable green concrete containing waste foundry sand

The use of waste foundry sand (WFS) in concrete production has gained attention as an eco-friendly approach to waste reduction and enhancing cementitious materials. However, testing the impact of WFS in concrete through experiments is costly and time-consuming. Therefore, this study employs machine learning (ML) models, including support vector regression (SVR), decision tree (DT), and AdaBoost regressor (AR) ensemble model to predict concrete properties accurately. Moreover, SVR was employed in conjunction with three robust optimization algorithms: the firefly algorithm (FFA), particle swarm optimization (PSO), and grey wolf optimization (GWO), to construct hybrid models. Using 397 experimental data points for compressive strength (CS), 146 for elastic modulus (E), and 242 for split tensile strength (STS), the models were evaluated with statistical metrics and interpreted using the SHapley Additive exPlanation (SHAP) technique. The SVR-GWO hybrid model demonstrated exceptional accuracy in predicting waste foundry sand concrete (WFSC) strength characteristics. The SVR-GWO hybrid model exhibited correlation coefficient values (R) of 0.999 for CS and E, and 0.998 for STS. Age was found to be a significant factor influencing WFSC properties. The ensemble model (AR) also exhibited comparable prediction accuracy to the SVR-GWO model. In addition, SHAP analysis revealed an optimal content of input variables in the concrete mix. Overall, the hybrid and ensemble models showed exceptional prediction accuracy compared to individual models. The application of these sophisticated soft computing prediction techniques holds the potential to stimulate the widespread adoption of WFS in sustainable concrete production, thereby fostering waste reduction and bolstering the adoption of environmentally conscious construction practices.

Low-Carbon Operation Strategy of Park-Level Integrated Energy System with Firefly Algorithm

The integrated energy system at the park level, renowned for its diverse energy complementarity and environmentally friendly attributes, serves as a crucial platform for incorporating novel energy consumption methods. Nevertheless, distributed energy generation, characterized by randomness, fluctuations, and intermittency, is significantly influenced by the surrounding environment. Within the park, the output of multiple devices frequently diverges significantly from the actual demand, potentially resulting in energy waste phenomena, such as the curtailment of wind and solar power. To tackle the dual challenges of balancing energy supply and demand while reducing carbon emissions in the industrial park, this paper introduces a low-carbon integrated energy system that incorporates distributed renewable and clean energy sources. Mathematical models are formulated for the source–grid–load–storage components of this low-carbon integrated energy system. Furthermore, various operational scenarios for the park-level integrated energy system are analyzed. The ultimate goal is to devise an economically viable, low-carbon, and efficient operational strategy for the integrated energy system, aiming to satisfy the diverse objectives of various stakeholders.

Adaptive firefly algorithm for resource allocation and modified advanced encryption standard algorithm for hypervisor attack detection on cloud computing

Introduction: The advantages and wide-ranging applications of cloud computing have made it a prominent attention for scholars these days. The dispersed structure of cloud and its whole dependence on the internet for service delivery extant security problems. Methods: Hypervisor attack detection is carried out in this study using Modified Advanced Encryption Standard (MAES) system which ensures security prominently. It finds and detects the attacks earlier for secured VM migration. The proposed system includes main phases including system framework, load balancing, resource allocation and hypervisor attack detection via MAES algorithm. Initially, Over the duration of cloud computing, consider the quantity of tasks, VM, and cloud users. In this research, the MMH system is employed for load balancing to balance the total burden across the cloud. Tasks are moved from overloaded to underloaded nodes to attain load balancing. Next, the allocation of resources is carried out utilizing Adaptive Firefly (AF) optimization system which is used to select best resources optimally. It generates the best fitness values to choose the best resources. Results: It is also focused to improve the cost metric, computational complexity, throughput and VM performance in cloud. Then, to detect hypervisor attacks, the MAES method is employed. It specializes on offering enhanced security for cloud data and is employed to identify hypervisor and VM attackers. Conclusion: The findings produced the conclusion that the suggested MAES method superior to the current approaches according to throughput, computation cost, Mean Square Error (MSE) rate, and energy use.

Cooperative task allocation method for air-sea heterogeneous unmanned system with an application to ocean environment information monitoring

Ocean environment information monitoring is crucial to maritime battlefield situation awareness, marine disaster prevention, and marine resources utilization. However, current monitoring means, including remote sensing satellites, scientific investigation ships, and ocean observation stations, have limitations in terms of spatial coverage, spatiotemporal resolution, and manpower resources. Therefore, this study proposed the Air-Sea heterogeneous unmanned system composed of UAVs and USVs to carry out ocean environment information monitoring tasks.) Firstly, an algorithm named FASO (Firefly Algorithm Based Particle Swarm Optimization) is proposed to obtain the deployment of monitoring nodes in the sea-surface and airspace. Secondly, a multi-objective optimization cooperative task allocation model is established, which contains complex constraints and takes the energy consumption cost of UAVs and USVs as optimization goals. Finally, a cooperative task allocation method based on the improved NSGA-II algorithm is presented to enhance the convergence speed and quality of the cooperative task allocation for air-sea heterogeneous unmanned system. Simulation results demonstrate that the proposed algorithm have higher generality and superiority compared to other algorithms. This study provides a specific approach for the application of heterogeneous unmanned systems in the field of ocean monitoring.

Enhancing wireless sensor network connectivity and coverage using Hybrid GWO‐HSA algorithm

SummaryWireless sensor networks (WSNs) are essential in environmental monitoring, healthcare, and industrial automation. Persistent connectivity and coverage challenges in WSN stem from intermittent node connectivity due to obstacles, signal interference, node failures, and compromised data reliability. Existing solutions, while useful, exhibit limitations in fully addressing these concerns. To confront these challenges, a proposed system introduces the Hybrid Grey Optimizer–Harmony Search Algorithm (Hybrid GWO‐HSA), merging adaptive routing protocols and efficient deployment techniques. The Hybrid GWO‐HSA system conducts an initial environmental analysis to pinpoint factors affecting node communication. It strategically deploys additional nodes to bridge coverage gaps, using the Grey Wolf Algorithm's capabilities to optimize node placement. Moreover, it employs the Harmony Search Algorithm to dynamically adjust communication paths based on real‐time network conditions, ensuring robust data transmission. The system workflow involves an environmental assessment followed by node deployment guided by the Grey Wolf Algorithm. Subsequently, the Harmony Search adapts communication paths to enhance connectivity. Simulations and practical experiments across diverse environments validate the Hybrid GWO‐HSA system's effectiveness. Results showcase substantial improvements: network lifetime of 13,200 s, a network delay of 37 ms, a coverage rate of 0.88, and an energy consumption of 590 J. This Hybrid GWO‐HSA‐based system establishes a resilient and efficient WSN infrastructure vital for reliable data collection and transmission in challenging settings. The Hybrid GWO‐HSA system offers a comprehensive approach to WSN connectivity and coverage issues by leveraging firefly and genetic algorithms, and it significantly enhances WSN performance and reliability across multifaceted application domains.

Teaching Resource Recommendation of Online Sports Collaborative Learning Platform Based on Optimized K-means Algorithm

The online collaborative learning platform for physical education is an interactive and open physical education teaching mode. To improve students’ learning interest and efficiency, the online sports collaborative learning platform is designed. From the perspective of person-post matching, the role in the group is designed and the improved clustering algorithm is used to realize the grouping. The combination of the k-mean algorithm and the firefly algorithm is used to enhance the real-time and accuracy of learning resource recommendation. The outcomes demonstrated that the Firefly algorithm had obvious advantages in convergence speed and other aspects. Relative to the classical K-means algorithm and the Firefly algorithm, the average clustering accuracy of the presented algorithm was improved by 7.23 % as well as 2.18 %, and the average processing time was improved by 4.35 % and 2.26 %, respectively. In the dataset Iris, the average clustering accuracy and processing time were 91.29 and 8.65, respectively. The optimal, worst, and average values of the online collaborative learning platform on the ground of the firefly-optimized K-means algorithm were 0.3006, 3.2176, and 1.5234, respectively. The fusion algorithm proposed in this study can optimize the recommendation of teaching resources on sports online collaborative learning platforms, improve learners’ learning passion, learning efficiency, and satisfaction, and relieve teachers’ teaching pressure.