A systematic literature review on meta-heuristic based feature selection techniques for text classification.

Feature selection (FS) is a critical step in many data science-based applications, especially in text classification, as it includes selecting relevant and important features from an original feature set. This process can improve learning accuracy, streamline learning duration, and simplify outcomes. In text classification, there are often many excessive and unrelated features that impact performance of the applied classifiers, and various techniques have been suggested to tackle this problem, categorized as traditional techniques and meta-heuristic (MH) techniques. In order to discover the optimal subset of features, FS processes require a search strategy, and MH techniques use various strategies to strike a balance between exploration and exploitation. The goal of this research article is to systematically analyze the MH techniques used for FS between 2015 and 2022, focusing on 108 primary studies from three different databases such as Scopus, Science Direct, and Google Scholar to identify the techniques used, as well as their strengths and weaknesses. The findings indicate that MH techniques are efficient and outperform traditional techniques, with the potential for further exploration of MH techniques such as Ringed Seal Search (RSS) to improve FS in several applications.

A clustering mining method for sports behavior characteristics of athletes based on the ant colony optimization.

This study aims to enhance the precision of analyzing athlete behavior characteristics, thereby optimizing sports training and competitive strategies. This study introduces an innovative Ant Colony Optimization (ACO) clustering model designed to address the high-dimensional clustering issues in athlete behavior data by simulating the path selection mechanism of ants searching for food. The development process of this model includes fine-tuning ACO parameters, optimizing for features specific to sports data, and comparing it with traditional clustering algorithms, and similar research models based on the neural network, support vector machines, and deep learning. The results indicate that the ACO model significantly outperforms the comparison algorithms in terms of silhouette coefficient (0.72) and Davies-Bouldin index (1.05), demonstrating higher clustering effectiveness and model stability. Particularly noteworthy is the recall rate (0.82), a key performance indicator, where the ACO model accurately captures different behavioral characteristics of athletes, validating its effectiveness and reliability in athlete behavior analysis. The innovation lies not only in the application of the ACO algorithm to address practical issues in the field of sports but also in showcasing the advantages of the ACO algorithm in handling complex, high-dimensional sports data. However, its generality and efficiency on a larger scale or different types of sports data still need further validation. In conclusion, through the introduction and optimization of the ACO clustering model, this study provides a novel and effective approach for a deeper understanding and analysis of athlete behavior characteristics. This study holds significant importance in advancing sports science research and practical applications.

Predictive modeling of deep vein thrombosis risk in hospitalized patients: A Q-learning enhanced feature selection model.

Deep vein thrombosis (DVT) represents a critical health concern due to its potential to lead to pulmonary embolism, a life-threatening complication. Early identification and prediction of DVT are crucial to prevent thromboembolic events and implement timely prophylactic measures in high-risk individuals. This study aims to examine the risk determinants associated with acute lower extremity DVT in hospitalized individuals. Additionally, it introduces an innovative approach by integrating Q-learning augmented colony predation search ant colony optimizer (QL-CPSACO) into the analysis. This algorithm, then combined with support vector machines (SVM), forms a bQL-CPSACO-SVM feature selection model dedicated to crafting a clinical risk prognostication model for DVT. The effectiveness of the proposed algorithm's optimization and the model's accuracy are assessed through experiments utilizing the CEC 2017 benchmark functions and predictive analyses on the DVT dataset. The experimental results reveal that the proposed model achieves an outstanding accuracy of 95.90% in predicting DVT. Key parameters such as D-dimer, normal plasma prothrombin time, prothrombin percentage activity, age, previously documented DVT, leukocyte count, and thrombocyte count demonstrate significant value in the prognostication of DVT. The proposed method provides a basis for risk assessment at the time of patient admission and offers substantial guidance to physicians in making therapeutic decisions.

Improved ant colony optimization for safe path planning of AUV.

In order to address the autonomous underwater vehicle navigation challenge for dam inspections, with the goal of enabling safe inspections and reliable obstacle avoidance, an improved smooth Ant Colony Optimization algorithm is proposed for path planning. This improved algorithm would optimize the smoothness of the path besides the robustness, avoidance of local optima, and fast computation speed. To achieve the goal of reducing turning time and improving the directional effect of path selection, a corner-turning heuristic function is introduced. Experimental simulation results show that the improved algorithm performs best than other algorithms in terms of path smoothness and iteration stability in path planning.

A new path planning strategy integrating improved ACO and DWA algorithms for mobile robots in dynamic environments.

This article is concerned with the path planning of mobile robots in dynamic environments. A new path planning strategy is proposed by integrating the improved ant colony optimization (ACO) and dynamic window approach (DWA) algorithms. An improved ACO is developed to produce a globally optimal path for mobile robots in static environments. Through improvements in the initialization of pheromones, heuristic function, and updating of pheromones, the improved ACO can lead to a shorter path with fewer turning points in fewer iterations. Based on the globally optimal path, a modified DWA is presented for the path planning of mobile robots in dynamic environments. By deleting the redundant nodes, optimizing the initial orientation, and improving the evaluation function, the modified DWA can result in a more efficient path for mobile robots to avoid moving obstacles. Some simulations are conducted in different environments, which confirm the effectiveness and superiority of the proposed path planning algorithms.

A novel parallel ant colony optimization algorithm for mobile robot path planning.

With the continuous development of mobile robot technology, its application fields are becoming increasingly widespread, and path planning is one of the most important topics in the field of mobile robot research. This paper focused on the study of the path planning problem for mobile robots in a complex environment based on the ant colony optimization (ACO) algorithm. In order to solve the problems of local optimum, susceptibility to deadlocks, and low search efficiency in the traditional ACO algorithm, a novel parallel ACO (PACO) algorithm was proposed. The algorithm constructed a rank-based pheromone updating method to balance exploration space and convergence speed and introduced a hybrid strategy of continuing to work and killing directly to address the problem of deadlocks. Furthermore, in order to efficiently realize the path planning in complex environments, the algorithm first found a better location for decomposing the original problem into two subproblems and then solved them using a parallel programming method-single program multiple data (SPMD)-in MATLAB. In different grid map environments, simulation experiments were carried out. The experimental results showed that on grid maps with scales of 20 $ \times $ 20, 30 $ \times $ 30, and 40 $ \times $ 40 compared to nonparallel ACO algorithms, the proposed PACO algorithm had less loss of solution accuracy but reduced the average total time by 50.71, 46.83 and 46.03%, respectively, demonstrating good solution performance.

Statistical versus neural network-embedded swarm intelligence optimization of a metallo-neutral-protease production: activity kinetics and food industry applications.

An integrated approach involving response surface methodology (RSM) and artificial neural network-ant-colony hybrid optimization (ANN-ACO) was adopted to develop a bioprocess medium to increase the yield of Bacillus cereus neutral protease under submerged fermentation conditions. The ANN-ACO model was comparatively superior (predicted r2 = 98.5%, mean squared error [MSE] = 0.0353) to RSM model (predicted r2 = 86.4%, MSE = 23.85) in predictive capability arising from its low performance error. The hybrid model recommended a medium containing (gL-1) molasses 45.00, urea 9.81, casein 25.45, Ca2+ 1.23, Zn2+ 0.021, Mn2+ 0.020, and 4.45% (vv-1) inoculum, for a 6.75-fold increase in protease activity from a baseline of 76.63 UmL-1. Yield was further increased in a 5-L bioreactor to a final volumetric productivity of 3.472 mg(Lh)-1. The 10.0-fold purified 46.6-kDa-enzyme had maximum activity at pH 6.5, 45-55 °C, with Km of 6.92 mM, Vmax of 769.23 µmolmL-1 min-1, kcat of 28.49 s-1, and kcat/Km of 4.117 × 103 M-1 s-1, at 45 °C, pH 6.5. The enzyme was stabilized by Ca2+, activated by Zn2+ but inhibited by EDTA suggesting that it was a metallo-protease. The biomolecule significantly clarified orange and pineapple juices indicating its food industry application.

Encapsulation of quercetin fraction from Musa balbisiana banana blossom in chitosan alginate solution, its optimization and characterizations.

This study comprises the isolation of quercetin from the bhimkol banana (Musa balbisiana) blossom, encapsulation, and its characterizations. An isolated quercetin rich fraction was obtained from HPLC followed by column chromatography and subsequently encapsulated with chitosan-alginate polyelectrolyte complex at optimum encapsulation conditions obtained by ant colony optimization. Quercetin fraction and encapsulated quercetin were characterized for their physicochemical properties (by HPLC, FTIR, NMR, XRD, Dynamic Light Scattering, and release study). The yield and purity of isolated quercetin rich fractions were 2.35 ± 0.08 µg/ml and 83.12 ± 0.31 %, respectively. After the optimization of encapsulation, quercetin 0.2 %, sodium alginate 4 %, chitosan 0.5 %, and agitation at 300 rpm were found to be the optimal conditions resulting in higher encapsulation efficiency (EE, 84.54 %). EE was significantly improved by a slight increase in sodium alginate, and agitation. Encapsulated quercetin revealed good pH resistance by releasing 68.27 mg QE/g quercetin in simulated gastric fluid at 60 min. Microbeads of encapsulated quercetin showed the structural bond stretching of encapsulating materials and quercetin in FTIR spectra (stretching at 1511 cm-1, 1380 cm-1, and 1241 cm-1 are attributed to the stretching vibration of CO in aromatic rings, and bending vibration of OH bond in phenols). An average particle size of 2.71 µm exhibited the microgel behavior of microbeads (by XRD). The present study on the underutilized variety of banana blossoms has diverse applications in the food and pharmaceutical industries that will productively exhibit effective drug delivery properties.

Adaptive Ant Colony Optimization with Sub-Population and Fuzzy Logic for 3D Laser Scanning Path Planning.

For the precise measurement of complex surfaces, determining the position, direction, and path of a laser sensor probe is crucial before obtaining exact measurements. Accurate surface measurement hinges on modifying the overtures of a laser sensor and planning the scan path of the point laser displacement sensor probe to optimize the alignment of its measurement velocity and accuracy. This manuscript proposes a 3D surface laser scanning path planning technique that utilizes adaptive ant colony optimization with sub-population and fuzzy logic (SFACO), which involves the consideration of the measurement point layout, probe attitude, and path planning. Firstly, this study is based on a four-coordinate measuring machine paired with a point laser displacement sensor probe. The laser scanning four-coordinate measuring instrument is used to establish a coordinate system, and the relationship between them is transformed. The readings of each axis of the object being measured under the normal measuring attitude are then reversed through the coordinate system transformation, thus resulting in the optimal measuring attitude. The nominal distance matrix, which demonstrates the significance of the optimal measuring attitude, is then created based on the readings of all the points to be measured. Subsequently, a fuzzy ACO algorithm that integrates multiple swarm adaptive and dynamic domain structures is suggested to enhance the algorithm's performance by refining and utilizing multiple swarm adaptive and fuzzy operators. The efficacy of the algorithm is verified through experiments with 13 popular TSP benchmark datasets, thereby demonstrating the complexity of the SFACO approach. Ultimately, the path planning problem of surface 3D laser scanning measurement is addressed by employing the proposed SFACO algorithm in conjunction with a nominal distance matrix.

Ant colony optimization for parallel test assembly.

Ant colony optimization (ACO) algorithms have previously been used to compile single short scales of psychological constructs. In the present article, we showcase the versatility of the ACO to construct multiple parallel short scales that adhere to several competing and interacting criteria simultaneously. Based on an initial pool of 120 knowledge items, we assembled three 12-item tests that (a) adequately cover the construct at the domain level, (b) follow a unidimensional measurement model, (c) allow reliable and (d) precise measurement of factual knowledge, and (e) are gender-fair. Moreover, we aligned the test characteristic and test information functions of the three tests to establish the equivalence of the tests. We cross-validated the assembled short scales and investigated their association with the full scale and covariates that were not included in the optimization procedure. Finally, we discuss potential extensions to metaheuristic test assembly and the equivalence of parallel knowledge tests in general.

Task Assignment and Path Planning Mechanism Based on Grade-Matching Degree and Task Similarity in Participatory Crowdsensing.

Participatory crowdsensing (PCS) is an innovative data sensing paradigm that leverages the sensors carried in mobile devices to collect large-scale environmental information and personal behavioral data with the user's participation. In PCS, task assignment and path planning pose complex challenges. Previous studies have only focused on the assignment of individual tasks, neglecting or overlooking the associations between tasks. In practice, users often tend to execute similar tasks when choosing assignments. Additionally, users frequently engage in tasks that do not match their abilities, leading to poor task quality or resource wastage. This paper introduces a multi-task assignment and path-planning problem (MTAPP), which defines utility as the ratio of a user's profit to the time spent on task execution. The optimization goal of MATPP is to maximize the utility of all users in the context of task assignment, allocate a set of task locations to a group of workers, and generate execution paths. To solve the MATPP, this study proposes a grade-matching degree and similarity-based mechanism (GSBM) in which the grade-matching degree determines the user's income. It also establishes a mathematical model, based on similarity, to investigate the impact of task similarity on user task completion. Finally, an improved ant colony optimization (IACO) algorithm, combining the ant colony and greedy algorithms, is employed to maximize total utility. The simulation results demonstrate its superior performance in terms of task coverage, average task completion rate, user profits, and task assignment rationality compared to other algorithms.

Finding the Link Between Iranian EFL Teacher Motivation and Engagement via Ant Colony Optimization Algorithm and Fuzzy Decision Mode.

Teacher motivation is considred as one of the most decisive factorts infulencing teacher functioing as well as students' achievement. Many variable can develop teacher motoivation. In this study, it is presumed that teacher engagement, comprising three facets of emotional, behavioral, and cognitive influence teacher motivation. To examine this hypothesis, this study takes the initiative to utiliuze an innovative artificial intelliengce (AI)-inspired approach called Ant Colony Optimization (ACO) technique. ACO is an artificial intelligence (AI) algorithm originating from natural phenomena. The concept originates from biology and physics and specifically from ants' movements. ACO has the ability to find the connections between inputs and outputs, and it can find the most influencing inputs. Motivation was the output of the study, and the inputs were three different engagement factors. Based on the results, ACO reached a high R-value meaning that it could predict the output with a high accuracy. The findings of this study substantiate the wide-ranging and multifacsted potentials of AI, in particular ACO, in studying and predicting human functioning in academic settings.

Optimizing the extraction of active components from Salvia miltiorrhiza by combination of machine learning models and intelligent optimization algorithms and its correlation analysis of antioxidant activity.

We extracted Sal B and TIIA from Salvia miltiorrhiza using enzymatic-assisted ethanol extraction. ACONN predicted optimal process conditions. Enzymolysis and alcohol extraction were used, optimizing conditions and evaluating antioxidant activity. ACONN analyzed data and ACO optimized conditions. Lab verification comprehensively evaluated the conditions. The correlation between Sal B, TIIA, and their antioxidant activities was examined. Weights of 0.5739 and 0.4260 evaluated Sal B and TIIA. ACONN had a 97.46% fitting degree. Optimized extraction conditions improved yield and quality, yielding a comprehensive evaluation value of 27.69 with 4.46% average errors. This approach enhances extraction and compound quality. Antioxidant activity strongly correlated with component yield, influenced by extraction conditions. ACONN-optimized extraction improved Sal B and TIIA yield and quality, with potential as natural antioxidants. Integrating machine learning and optimization algorithms in industrial extraction enhances efficiency and environmental preservation.

The 10-Item Short Form of the German Experiences in Close Relationships Scale (ECR-G-10)-Model Fit, Reliability, and Validity.

The aim of the present work was the development and validation of a short form of the Experiences in Close Relationship Scale (ECR) in German. Three studies were conducted. In study 1, the best items for the short form were selected from the item pool of the original version based on ant colony optimization (ACO), a recently developed probabilistic approach. Data from three samples collected at a university, an online portal, and a psychosomatic clinic with a total of 1470 participants were analyzed. A 10-item solution resulted, measuring avoidance and anxiety with five items each. This solution showed a good model fit and acceptable reliability in all three samples. The two new short scales were independent of each other. In study 2, the 10-item solution was validated by correlating the new short scales with external criteria. Data from previous studies that included student, community, and clinical samples were reanalyzed. Both short scales showed expected correlations with measures of romantic relationships, personality, psychopathology, and childhood trauma, indicating convergent and discriminant validity. The significant correlations were moderate to strong. In study 3, the selected ten items alone and several content-related scales were presented online to 277 participants, most of them students. The good results in terms of model fit, reliability, and validity observed in studies 1 and 2 could be replicated here. The new short form, called ECR-G-10, allows the measurement of attachment avoidance and anxiety in an economic way in research and clinical practice.

Analysis of computational intelligence approaches for predicting disease severity in humans: Challenges and research guidelines.

The word disease is a common word and there are many diseases like heart disease, diabetes, breast cancer, COVID-19, and kidney disease that threaten humans. Data-mining methods are proving to be increasingly beneficial in the present day, especially in the field of medical applications; through the use of machine-learning methods, that are used to extract valuable information from healthcare data, which can then be used to predict and treat diseases early, reducing the risk of human life. Machine-learning techniques are useful especially in the field of health care in extracting information from healthcare data. These data are very much helpful in predicting the disease early and treating the patients to reduce the risk of human life. For classification and decision-making, data mining is very much suitable. In this paper, a comprehensive study on several diseases and diverse machine-learning approaches that are functional to predict those diseases and also the different datasets used in prediction and making decisions are discussed in detail. The drawbacks of the models from various research papers have been observed and reveal countless computational intelligence approaches. Naïve Bayes, logistic regression (LR), SVM, and random forest are able to produce the best accuracy. With further optimization algorithms like genetic algorithm, particle swarm optimization, and ant colony optimization combined with machine learning, better performance can be achieved in terms of accuracy, specificity, precision, recall, and specificity.

Optimizing Microsatellite Marker Panels for Genetic Diversity and Population Genetic Studies: An Ant Colony Algorithm Approach with Polymorphic Information Content.

Microsatellites are polymorphic and cost-effective. Optimizing reduced microsatellite panels using heuristic algorithms eases budget constraints in genetic diversity and population genetic assessments. Microsatellite marker efficiency is strongly associated with its polymorphism and is quantified as the polymorphic information content (PIC). Nevertheless, marker selection cannot rely solely on PIC. In this study, the ant colony optimization (ACO) algorithm, a widely recognized optimization method, was adopted to create an enhanced selection scheme for refining microsatellite marker panels, called the PIC-ACO selection scheme. The algorithm was fine-tuned and validated using extensive datasets of chicken (Gallus gallus) and Chinese gorals (Naemorhedus griseus) from our previous studies. In contrast to basic optimization algorithms that stochastically initialize potential outputs, our selection algorithm utilizes the PIC values of markers to prime the ACO process. This increases the global solution discovery speed while reducing the likelihood of becoming trapped in local solutions. This process facilitated the acquisition of a cost-efficient and optimized microsatellite marker panel for studying genetic diversity and population genetic datasets. The established microsatellite efficiency metrics such as PIC, allele richness, and heterozygosity were correlated with the actual effectiveness of the microsatellite marker panel. This approach could substantially reduce budgetary barriers to population genetic assessments, breeding, and conservation programs.

Optimal layout design of groundwater pollution monitoring network using parameter iterative updating strategy-based ant colony optimization algorithm.

The scientific layout design of the groundwater pollution monitoring network (GPMN) can provide high quality groundwater monitoring data, which is essential for the timely detection and remediation of groundwater pollution. The simulation optimization approach was effective in obtaining the optimal design of the GPMN. The ant colony optimization (ACO) algorithm is an effective method for solving optimization models. However, the parameters used in the conventional ACO algorithm are empirically adopted with fixed values, which may affect the global searchability and convergence speed. Therefore, a parameter-iterative updating strategy-based ant colony optimization (PIUSACO) algorithm was proposed to solve this problem. For the GPMN optimal design problem, a simulation-optimization framework using PIUSACO algorithm was applied in a municipal waste landfill in BaiCheng city in China. Moreover, to reduce the computational load of the design process while considering the uncertainty of aquifer parameters and pollution sources, a genetic algorithm-support vector regression (GA-SVR) method was proposed to develop the surrogate model for the numerical model. The results showed that the layout scheme obtained using the PIUSACO algorithm had a significantly higher detection rate than ACO algorithm and random layout schemes, indicating that the designed layout scheme based on the PIUSACO algorithm can detect the groundwater pollution occurrence timely. The comparison of the iteration processes of the PIUSACO and conventional ACO algorithms shows that the global searching ability is improved and the convergence speed is accelerated significantly using the iteration updating strategy of crucial parameters. This study demonstrates the feasibility of the PIUSACO algorithm for the optimal layout design of the GPMN for the timely detection of groundwater pollution.

Designing a sustainable collection and transportation routes for domestic wastes in the agro-pastoral ecotone of the Tibetan Plateau.

The absence of an efficient and safe routes for the timely collection and transportation of domestic waste (DW) may have negative effects on the environment and public health. However, the existing collection and transportation routes (CTR) for domestic waste (DW) based on territorial management are not suitable for the special socio-ecological system of the agro-pastoral ecotone (APE). Therefore, it is crucial to develop a low-cost, high-efficiency, and risk-free CTR to mitigate the threat of DW to the environmental sustainability in the APE of the Tibetan Plateau. This study selected Haidong as a research case and constructed a sustainable CTR optimization framework based on an integrated perspective on temporal, spatial and eco-safety risk. We used the improved Ant Colony Optimization (ACO) to simulate optimal spatial-temporal routes, and the eco-safety risk level of the CTR was assessed by using the Minimum Cumulative Resistance model (MCR). Results demonstrated that: (1) After the sustainable model was optimized, the total transportation mileage and the frequency of collection and transportation were reduced by 45.88% and 38.07% respectively, the economic cost savings were decreased by 32.29%. Optimized routes were more effective and can better adapt to the dispersed pollution-producing characteristics in the APE. (2) The optimized routes reduced greenhouse gas (GHG) emissions by 41.09%, and reduced the eco-safety risk of the high and relative high-risk routes, which account for 29.05% of total routes, can protect important ecological functions and reduce the adverse impacts of DW transportation on soil, atmosphere, water, and the living environment. (3) The cores of adaptive management for sustainable CTR in APE were to change from the current single-county administrative organization to a cross-county administrative organization; adjust the transportation cycle based on pollution-producing characteristics; sort the DW locally; and cultivate environmental awareness among farmers and herdsmen. This study designed new sustainable collection and transportation routes for domestic waste to improve environmental sustainability in the agro-pastoral ecotone.

Hierarchical Area-Based and Path-Based Heuristic Approaches for Multirobot Coverage Path Planning with Performance Analysis in Surveillance Systems.

In this study, we present a systematic exploration of hierarchical designs for multirobot coverage path planning (MCPP) with a special focus on surveillance applications. Unlike conventional studies centered on cleaning tasks, our investigation delves into the realm of surveillance problems, specifically incorporating the sensing range (SR) factor equipped on the robots. Conventional path-based MCPP strategies considering SR, primarily rely on naive approaches, generating nodes (viewpoints) to be visited and a global path based on these nodes. Therefore, our study proposes a general MCPP framework for surveillance by dealing with path-based and area-based structures comprehensively. As the traveling salesman problem (TSP) solvers, our approach incorporates not the naive approach but renowned and powerful algorithms such as genetic algorithms (GAs), and ant colony optimization (ACO) to enhance the planning process. We devise six distinct methods within the proposed MCPP framework. Two methods adopt area-based approaches which segments areas via a hierarchical max-flow routing algorithm based on SR and the number of robots. TSP challenges within each area are tackled using a GA or ACO, and the result paths are allocated to individual robots. The remaining four methods are categorized by the path-based approaches with global-local structures such as GA-GA, GA-ACO, ACO-GA, and ACO-ACO. Unlike conventional methods which requires a global path, we further incorporate ACO- or GA-based local planning. This supplementary step at the local level enhances the quality of the path-planning results, particularly when dealing with a large number of nodes, by preventing any degradation in global path-planning outcomes. An extensive comparative analysis is conducted to evaluate the proposed framework based on execution time, total path length, and idle time. The area-based approaches tend to show a better execution time and overall path length performance compared to the path-based approaches. However, the path-based MCPP methods have the advantage of having a smaller idle time than the area-based MCPP methods. Our study finds that the proposed area-based MCPP method excels in path planning, while the proposed path-based MCPP method demonstrates superior coverage balance performance. By selecting an appropriate MCPP structure based on the specific application requirements, leveraging the strengths of both methodologies, efficient MCPP execution becomes attainable. Looking forward, our future work will focus on tailoring these MCPP structures to diverse real-world conditions, aiming to propose the most suitable approach for specific applications.

Breast Cancer Classification Using Synthesized Deep Learning Model with Metaheuristic Optimization Algorithm.

Breast cancer is the second leading cause of mortality among women. Early and accurate detection plays a crucial role in lowering its mortality rate. Timely detection and classification of breast cancer enable the most effective treatment. Convolutional neural networks (CNNs) have significantly improved the accuracy of tumor detection and classification in medical imaging compared to traditional methods. This study proposes a comprehensive classification technique for identifying breast cancer, utilizing a synthesized CNN, an enhanced optimization algorithm, and transfer learning. The primary goal is to assist radiologists in rapidly identifying anomalies. To overcome inherent limitations, we modified the Ant Colony Optimization (ACO) technique with opposition-based learning (OBL). The Enhanced Ant Colony Optimization (EACO) methodology was then employed to determine the optimal hyperparameter values for the CNN architecture. Our proposed framework combines the Residual Network-101 (ResNet101) CNN architecture with the EACO algorithm, resulting in a new model dubbed EACO-ResNet101. Experimental analysis was conducted on the MIAS and DDSM (CBIS-DDSM) mammographic datasets. Compared to conventional methods, our proposed model achieved an impressive accuracy of 98.63%, sensitivity of 98.76%, and specificity of 98.89% on the CBIS-DDSM dataset. On the MIAS dataset, the proposed model achieved a classification accuracy of 99.15%, a sensitivity of 97.86%, and a specificity of 98.88%. These results demonstrate the superiority of the proposed EACO-ResNet101 over current methodologies.