Abstract: This paper introduces the Operational Coherence Framework (OCOF) v1.3, a formal architecture specifying the structural prerequisites for semantic interpretation in intelligent systems. The framework defines interpretive intelligence not through scale or behavioral sophistication, but through five independent operational axioms: Boundary Integrity, Precision Structuring, Semantic Valuation, Policy Alignment, and Global State Continuity. Each axiom imposes a distinct informational constraint, and their joint satisfaction delineates the operational envelope within which internal states can support meaningful structure. Rather than adopting emergent or capacity-based accounts of meaning, OCOF characterizes meaning as a condition of structural readiness—a phase transition that occurs only when boundary stability, signal reliability, valuation structure, action coherence, and temporal continuity collectively reach their coherence thresholds. The framework situates mechanisms from the Free Energy Principle, Predictive Processing, Integrated Information Theory, and Control Theory within this unified constraint architecture, showing that these models operate as specialized components presupposing the structural conditions defined by OCOF. A central contribution of this work is the operational definition of Meaning-Readiness, the point at which a system’s boundary integrity and precision structure allow the reliable attribution of semantic relevance beyond syntactic or associative processing. We demonstrate the logical independence and non-circularity of the five axioms, establishing OCOF as a self-contained and falsifiable theoretical kernel. As a result, OCOF v1.3 provides a substrate-neutral foundation for evaluating interpretive capacity in biological, artificial, and hybrid systems, offering a principled basis for cognitive modeling and AGI alignment.

Abstract: The growing complexity of BIM (Building Information Model) models leads to perfor-mance issues, extended file loading times, and difficulties in cross-industry coordina-tion. One of the main factors reducing performance are so-called "heavy" library com-ponents (families in Revit), characterized by excessive geometric complexity, a large number of instances, or improper optimization. Currently, the identification of such components is based mainly on the experience of designers and manual inspection of models, which is time-consuming and prone to errors. This article presents a new tool, HeavyFamilies, which automates the detection and analysis of heavy library compo-nents in BIM models. The tool uses a multi-criteria analysis method, evaluating com-ponents based on five key parameters: number of instances, geometry complexity, number of walls and edges, and estimated file size. Each parameter is weighed ac-cording to its impact on model performance. The developed solution has been imple-mented as a pyRevit plugin for Autodesk Revit, offering a graphical interface with a tabular summary of results, a CSV export function, and visualization of detected components directly in the model. Validation of the tool on real BIM projects has demonstrated its effectiveness in identifying components with a weight index exceed-ing the threshold of 200, allowing designers to prioritize optimization efforts. The HeavyFamilies tool is a practical contribution to the field of BIM model optimization, enabling a systematic approach to managing model performance in complex construc-tion projects and supporting the development of smart cities.

Abstract: Automation of the plant-based meat extrusion process requires a scheme to provide quantitative estimates of output fibrosity, which must be carried out online and in real-time. A novel machine learning regression model for this purpose, is proposed in this article. A deep neural network that was originally trained for image classification, was extended to provide quantitative fibrosity estimates. Relevant layers of the network were retrained using real-world laboratory data. Plant-based meat or textured vegetable protein products with varying fibrous microstructures were obtained using different ingredient formulations and process conditions on a pilot-scale twin screw extruder with in-barrel moisture range of 29.2-40.9% (wet basis). Images of extruded plant-based meat products were collected to serve as sample inputs. An experiment was devised, where image samples were randomly presented to two expert human subjects, who provided as feedback, fibrosity scores lying within the interval [1, 10]. Statistical metrics were adopted to evaluate the performance of the trained network. It was found that the network performed significantly better when trained separately with feedback scores of each individual subject, than with the combined scores, indicating that it was able to capture nuances of a subject’s perception. Another study was directed at the explainability of the network’s estimations. Using standard software, a set of synthetic images of varying shapes and sizes were created as inputs to the network. Interpretations of its output scores indicate that the network’s estimates were based on features relevant to porosity and fibrosity, while not influenced by extraneous ones.

Abstract: Knowledge of the maximum gust expected over a period of years is essential for off-shore structures design. Because long records of gust speed are not normally availa-ble, maximum gusts have traditionally been estimated by multiplying the maximum expected hourly or 10-minute wind speed by a gust factor. That calculation ignores the possibility that the highest gust might not occur in the hour with the highest mean wind speed. A similar problem arises in the estimation of the maximum expected in-dividual wave height. By analogy with the accepted method of calculating maximum wave heights, we demonstrate how maximum gusts can be calculated from time series of average wind speed and wind gust distributions. We used measurements from the IJmuiden meteorological mast offshore The Netherlands to find wind gust distribu-tions. The IJmuiden data is particularly useful for studying gusts because four years of measurements were made at a sampling frequency of 4 Hz. Those distributions were used to predict extreme values of gusts in a storm using methods similar to those used in wave height calculations. The resulting extreme values closely matched ex-treme values calculated directly from the measured maximum gusts in each storm. The methods described here can calculate extreme gust speeds more accurately than the methods currently in use.

Abstract: Post-disaster reconstruction remains largely excluded from circular-economy ap-proaches. This gap is particularly evident in earthquake-affected inner territories, where reconstruction faces severe logistical constraints—especially rubble manage-ment—and where debris is often composed of materials closely tied to local building cultures and community identities. In these contexts, rebuilding still follows linear, emergency-driven models that treat rubble primarily as waste. This study introduces Rubble as a Material Bank (RMB), a digital–material framework that reconceptualises earthquake rubble as a traceable and programmable resource for circular reconstruc-tion. RMB defines a rubble-to-component chain integrating material characterisation, data-driven management, robotic fabrication, and reversible architectural design. Se-lected downstream segments are experimentally validated through the TRAP project, developed within the European TARGET-X program. The experimentation focuses on extrusion-based fabrication of dry-assembled wall components using rubble-derived aggregates. Results show that digitally governed workflows can enable material reuse while revealing technical and regulatory constraints on large-scale implementation.

Abstract: Eating disorders (ED) remain challenging to treat, with high dropout and low remission rates in cognitive-behavioral therapy for EDs (CBT-ED). Psilocybin treatment (PT) demonstrates therapeutic potential to enhance CBT-ED by exerting several neurobiological, psychological, and experiential effects (e.g., antidepressant, neuroplasticity, emotional openness) that are hypothesized to increase psychotherapeutic engagement, reduce dropout, and improve clinical outcomes. This article provides the first consolidation of existing theoretical evidence for PT/CBT-ED, proposes considerations for a con-current intervention protocol, and presents clinical and research considerations to empirically test its feasibility, safety, and efficacy. This line of inquiry is expected to advance the development of approaches that improve ED treatment outcomes and, more broadly, advance the study of psychedelics as tools to enhance evidence-based psychotherapy models.

Abstract: Amino acid derivatives, such as β-keto esters and pyrrolones, were used as nucleophiles in organocatalyzed Michael additions to nitroalkene acceptors, while fatty acid derivatives acted as both nucleophiles (β-keto esters) and electrophile (nitroalkene acceptor). Bifunctional noncovalent organocatalysts were employed as asymmetric organocatalysts. Twenty compounds – including fatty acid and amino acid derivatives, as well as fatty acid–amino acid conjugates – were prepared with enantioselectivities of up to 98% ee. All novel products were fully characterized. This research demonstrates the ease of assembling readily available fatty acid and amino acid building blocks under ambient conditions.

Abstract: Zika virus (ZIKV), a mosquito-borne flavivirus, is associated with congenital malformations and neuroinflammatory disorders, highlighting the need to identify host factors that shape infection outcomes. Macrophages, key targets and reservoirs of ZIKV, orchestrate both antiviral and inflammatory responses. Vitamin D (VitD) has emerged as a potent immunomodulator that enhances macrophage antimicrobial activity and regulates inflammation. To investigate how VitD shapes macrophage responses to ZIKV, we reanalyzed publicly available RNA-seq and miRNA-seq datasets from monocyte-derived mac-rophages (MDMs) of four donors, differentiated with or without VitD and subsequently infected with ZIKV. Differential expression analysis identified long non-coding RNAs (lncRNAs), microRNAs (miRNAs), and mRNAs, integrated into competing endogenous RNA (ceRNA) networks. In VitD-conditioned and ZIKV-infected MDMs, 65 lncRNAs and 23 miRNAs were significantly modulated. Notably, lncRNAs such as HSD11B1-AS1, Lnc-FOSL2, SPIRE-AS1, and PCAT7 were predicted to regulate immune and metabolic genes, including G0S2, FOSL2, PRELID3A, and FBP1. Among the miRNAs, let-7a and miR-494 were downregulated, while miR-146a, miR-708, and miR-378 were upregulated, all of which have been previously implicated in antiviral immunity. Functional enrichment analysis revealed pathways linked to metabolism, stress responses, and cell migration. ceRNA network analysis suggested that SOX2-OT and SLC9A3-AS1 may act as molecular sponges, modulating regulatory axes relevant to immune control and viral response. Despite limitations in sample size and experimental validation, this study provides an exploratory map of ncRNA–mRNA networks shaped by VitD during ZIKV infection, highlighting candidate molecules and pathways for further studies on host–virus interactions and VitD-mediated immune regulation.

Abstract: Background/Objectives: Diagnosing Rotator Cuff Tears (RCTs) via Magnetic Resonance Imaging (MRI) is clinically challenging due to complex 3D anatomy and significant in-terobserver variability. Traditional slice-centric Convolutional Neural Networks (CNNs) often fail to capture the necessary volumetric context for accurate grading. This study aims to develop and validate the Patient-Aware Vision Transformer (Pa-ViT), an explainable deep learning framework designed for the automated, patient-level classification of RCTs (Normal, Partial-Thickness, and Full-Thickness). Methods: A large-scale retrospective dataset comprising 2,447 T2-weighted coronal shoulder MRI examinations was utilized. The proposed Pa-ViT framework employs a Vision Transformer (ViT-Base) backbone within a Weakly-Supervised Multiple Instance Learning (MIL) paradigm to aggregate slice-level semantic features into a unified patient diagnosis. The model was trained using a weighted cross-entropy loss to address class imbalance and was benchmarked against widely used CNN architectures and traditional machine learning classifiers. Results: The Pa-ViT model achieved a high overall accuracy of 91% and a macro-averaged F1-score of 0.91, significantly outperforming the standard VGG-16 baseline (87%). Notably, the model demonstrated superior discriminative power for the challenging Partial-Thickness Tear class (ROC AUC: 0.903). Furthermore, Attention Rollout visualizations confirmed the model’s reliance on genuine anatomical features, such as the supraspinatus footprint, rather than artifacts. Conclusions: By effectively modeling long-range dependencies, the Pa-ViT framework provides a robust alternative to traditional CNNs. It offers a clinically viable, explainable decision support tool that enhances diagnostic sensitivity, particularly for subtle partial-thickness tears.

Abstract: Lipid nanoparticles (LNPs) are a critical structural element of modern mRNA therapeutics, including COVID‑19 modRNA vaccines. Each formulation is a multicomponent system in which the LNP serves not as a passive carrier but as an active, biointeractive entity whose ionizable lipids engage directly with cellular membranes. Current evidence from cellular, transcriptomic, and proteomic analyses indicates that LNPs, with or without active mRNA cargo, alter transcriptomic programs and protein expression. This suggests that, even during uptake and interaction with the membrane (transfection), the membrane serves as an initial site for inflammatory, detoxifying, and stress responses. Simultaneously, pathways involved in fat metabolism and detoxification are affected, such as the peroxisome proliferator-activated receptor γ (PPARγ) and cytochrome P450 (CYP) enzyme systems. We believe that the phosphatidylinositol (PI) cycle is the initial point for these disorders. This cycle regulates both organelle trafficking and membrane restructuring following endocytic processes, including macropinocytosis. When this cycle is disrupted, membrane restructuring and organelle dysfunction occur, triggering downstream signaling cascades such as nuclear factor kappa-B (NF- κB), mitogen-activated protein kinases (MAPKs), Janus kinase–signal transducer (JAK-STAT) pathways, and mechanistic target of rapamycin (mTOR) complexes. Transfection with LNPs may induce a systemic condition we call lipid-nanoparticle-driven membrane dysfunction (L‑DMD), where transfection results in broader dysregulation of cellular communication, stress response, and energy balance. This hypothesis-driven review offers a mechanistic foundation for understanding the diffuse, often enduring, biological effects observed after exposure to messenger RNA LNP formulations. It highlights a needed perspective at the intracellular level and within systems biology.

Abstract: Background: Early evidence from a nursing home in Yepes (Toledo, Spain) indicated that antihistamines combined with azithromycin prevented deaths and hospitalizations dur-ing the first COVID-19 wave. Subsequent data from the Consorci Sanitari de Terrassa (CST) showed that patients chronically taking antihistamines significantly reduced hos-pital admissions and mortality. However, a concerning rise in long COVID incidence (2–5%) after the third infection and a doubling of thrombosis rates in patients over 60 were observed. Objective: This study aimed to determine whether chronic antihistamine pre-scription is associated with a reduction in long COVID syndrome and thrombotic events. Methods: We analyzed anonymized data from the CST population (n=192,651 as of March 2025). Variables included age, gender, chronic antihistamine use, number of chronic treatments (nT), COVID-19 vaccination status, SARS-CoV-2 infection history, long COVID (LC) incidence, and aggregated thrombotic events. Odds ratios (OR) were calculated using chi-square tests. Results: The prevalence of LC increased progressively with successive in-fections in the non-antihistamine group. No significant differences were found with the antihistamine group, which presented no LC cases among the 52 patients with three documented infections. Thrombotic events were significantly less frequent in antihista-mine users with at least one chronic prescription (p< 0.0001). Conclusions: Results suggest a protective effect of antihistamines against thrombotic events. While confirmation via multicenter, randomized trials is needed, a pragmatic approach using antihistamines could be considered for symptomatic patients in the early stage of infection.

Abstract:

Sosnovsky’s hogweed (Heracleum sosnowskyi Manden.) is an invasive plant species widely distributed across Eastern Europe and Russia that poses a serious threat to human health due to its pronounced phototoxic properties. Contact with the plant sap, followed by exposure to ultraviolet radiation, frequently results in phytophotodermatitis characterized by erythema, blistering, ulceration, and long-lasting hyperpigmentation. The photochemical injuries are primarily attributed to highly oxygenated secondary metabolites, notably furanocoumarins, which act as potent photosensitizers and induce cellular and DNA damage upon UV activation. This review provides a comprehensive overview of the botanical distribution and invasiveness of H. sosnowskyi, the chemical composition of its biologically active metabolites, and the molecular mechanisms underlying hogweed-induced skin injuries. Particular emphasis is placed on the photochemical transformations of furanocoumarins, including psoralens and their photooxidation products, such as 1,2-dioxetanes, which generate reactive oxygen species and DNA crosslinks. In addition, the review discusses other compounds derived from hogweed biomass, including furan derivatives, aromatic compounds, fatty acids, sterols, and their oxidative products, which may contribute to phototoxic and cytotoxic effects. Clinical manifestations of hogweed burns, their classification, symptomatology, and current therapeutic approaches are critically analyzed, highlighting the lack of standardized treatment guidelines. By integrating chemical, biological, and clinical data, this review aims to elucidate the mechanisms of photochemical skin injury caused by H. sosnowskyi and to support the development of more effective preventive and therapeutic strategies.

Abstract: The main theme of present comprehensive review paper is the microwave-assisted heat-ing (MWH) developed in CNR SCITEC laboratories in Genoa. By modifying a domestic microwave, this technique has been used to prepare various innovative materials through synthesis, sintering, or heating (foaming or melting). These materials include inorganic compounds like superconductive magnesium diboride (MgB2), as well as organic and or-ganic-inorganic composite. The review highlights the significant improvements in en-ergy efficiency, time saving, material properties, and environmental sustainability achieved through these processes. Specific applications discussed include the rotational molding of polyethylene powders, sintering of hydroxyapatite-based scaffolds, and the preparation of cork composites for sound-absorbing panels, expanded polystyrene com-posites for building elements, and polyvinylidene fluoride piezoelectric compo-sites. Future potential applications and market demand for these technologies are also explored.

Abstract:

Background/Objectives: Bacterial biofilms formed by Escherichia coli pose a significant challenge in veterinary medicine due to their intrinsic resistance to antibiotics. Antimicrobial peptides (AMPs) represent a promising alternative. AMPs exert their bactericidal activity by binding to negatively charged phospholipids in bacterial membranes via electrostatic interactions, leading to membrane disruption and rapid cell lysis. Methods: In vitro assays included MIC determination, biofilm eradication testing (crystal violet, colony counts, CLSM), swimming motility, and EPS quantification. CRISPR/Cas9 was used to construct and complement a kduD mutant. A transposon mutagenesis library was screened for biofilm-defective mutants. In vivo, a murine excisional wound infection model was treated with CRAMP-34, with wound closure and bacterial burden monitored. Gene expression changes were analyzed via RT-qPCR. Results: The mouse-derived AMP (abbreviation CRAMP-34) effectively eradicates pre-formed biofilms of a clinically relevant, porcine-origin E.coli strain and promotes wound healing in a murine infection model. We conducted a genome-wide transposon mutagenesis screen, which identified kduD, as a critical gene for robust biofilm formation. Functional characterization revealed that kduD deletion drastically impairs flagellar motility and alters exopolysaccharide production, leading to defective biofilm architecture without affecting growth. Notably, the anti-biofilm activity of CRAMP-34 phenocopied aspects of the kduD deletion, including motility inhibition and transcriptional repression of a common set of biofilm-related genes. Conclusions: The research highlight CRAMP-34 as a potent anti-biofilm agent and unveil kduD as a previously unrecognized regulator of E.coli biofilms development, whose associated pathway is implicated in the mechanism of action of CRAMP-34.

Abstract:

Hydrometallurgical pretreatment of pyrite-bearing concentrates and tailings by hydrothermal interaction with Cu(II) solutions is a promising route for chemical beneficiation and mitigation of acid mine drainage but is limited by passivation caused by elemental sulfur and secondary copper sulfides. Here, the effect of sodium lignosulfonate (SLS) on the hydrothermal reaction between natural pyrite and CuSO₄ in H₂SO₄ media at 180–220 °C was studied at [H₂SO₄]₀ = 10–30 g/dm³, [Cu]₀ = 6–24 g/dm³ and [SLS]₀ = 0–1.0 g/dm³. Process efficiency was evaluated by Fe extraction into solution and Cu precipitation on the solid phase, and products were characterized by XRD and SEM/EDS. SLS markedly intensified pyrite conversion: at 200 °C and 120 min Fe extraction increased from 14 to 26 % and Cu precipitation from 5 to 23 %, while at 220 °C Fe extraction reached 33.4 % and Cu precipitation 26.8 %. XRD confirmed the sequential transformation CuS → Cu_1.8S. SEM/EDS showed that SLS converts localized nucleation of Cu_xS on defect sites into the formation of a fine, loosely packed and well-dispersed copper sulfide phase. The results demonstrate that lignosulfonate surfactants efficiently suppress passivation and enhance mass transfer, providing a basis for intensifying hydrothermal pretreatment of pyrite-bearing industrial materials.

Abstract:

This paper presents a method for describing the differential equations of motion of mechanical systems using the Kuznetsov tensor. Traditional approaches to solving equations of motion rely on vector and matrix methods, but the proposed approach allows for significant simplification and generalization of problems by using a system state tensor. The paper discusses the main principles of working with the Kuznetsov tensor, which describes the evolution of the system in a unified context. Specifically, it outlines a method for integrating the equations of motion for various mechanical systems, such as oscillations in a two-mass spring system. Conditions for damping oscillations and controlling amplitude are also considered, expanding the applicability of the Kuznetsov tensor in engineering calculations. The advantages of the proposed approach include a more compact representation of the system of equations, ease of analyzing invariants and symmetries, and the ability to apply the method to multi-linked and multi-component systems. The use of the Kuznetsov tensor for modeling the dynamics of various systems represents a step toward a more universal approach in mechanics and engineering.

Abstract:

Background: Midlife is a period of heightened vulnerability to menopausal symptoms and body image concerns. However, little is known about how the experience of menopausal symptoms relates to the awareness of and attention toward internal body signals. Taking a dimensional approach, this study employed network analysis to examine how menopausal symptom domains relate to dimensions of interoceptive sensibility and body image in middle-aged women and identified the most influential and bridging features within this interconnected system. Methods: Two hundred and thirteen cisgender women aged 40–60 years residing in Ireland completed online measures of body appreciation (BAS-2), state body satisfaction (BISS), interoceptive sensibility (MAIA-2), and menopausal symptoms (Menopause Rating Scale). Results: Attention Regulation, Trusting, Body Appreciation, and Body Listening showed the highest expected influence. Body Appreciation emerged as the strongest bridge node, connecting interoceptive sensibility, body image, and menopausal symptoms. Trusting was negatively associated with psychological symptoms, whereas Noticing was positively associated with somatic symptoms. Regression analyses showed that lower body appreciation predicted greater somatic, urogenital, and psychological symptom severity, and lower Trusting predicted higher psychological symptom severity. Older age was associated with higher somatic and urogenital symptoms, while younger age was associated with higher psychological symptoms. Conclusions: Findings suggest that body appreciation and interoceptive trust are central, bridging processes in women’s experience of menopausal symptoms. Interventions that enhance body appreciation and interoceptive trust may help reduce psychological and physical symptom burden during the menopausal transition.

Abstract:

Background: Apelin-36 may be used to identify patients with ST-segment elevation myocardial infarction (STEMI) who are at risk for the no-reflow phenomenon. Patients presenting with STEMI were evaluated and stratified according to their apelin-36 levels. Methods: In this study, 161 patients presenting with STEMI within 12 hours of symptom onset and undergoing primary percutaneous coronary intervention (pPCI) were enrolled. Biochemical parameters, including apelin-36, troponin T, creatine kinase (CK), the MB fraction of creatine kinase (CK-MB), total cholesterol, triglycerides, and other routine laboratory parameters, were measured. Blood samples for apelin-36 measurement were collected prior to PCI, centrifuged to obtain serum, and preserved at -80⁰C until being assayed. Two-dimensional echocardiography was performed in all patients. Thereafter, patients were divided into two groups according to their level of Apelin-36. Results: Among the 161 consecutive STEMI patients, 115 (71.42%) had Apelin-36 levels ≤0.58ng/mL (group 1), whereas 46 (28.57%) had Apelin-36 levels >0.58ng/mL (group 2). In total, 51 (31.67%) STEMI patients experienced no-reflow phenomenon after PCI: 29 (18.01%) patients with apelin-36 ≤ 0.58ng/mL and 22 (13.66%) with a value > 0.58ng/mL (p < 0.001). In terms of Gensini score, the mean value in group 1 was (70.29 (±28.76), while in group 2, it was 81.95 (±23.82) (p=0.004). Overall, a positive correlation between apelin-36 and Gensini score was observed in both groups using Kendall’s correlation analysis (group 1: Figure 2, p=0.05; group 2: Figure 2, p<0.0001). Binary logistic regression analysis identified apelin-36 and diabetes mellitus as significant predictors at the 5% level, with p-values of 0.045 and 0.036, respectively. Patients with apelin-36 levels ≤ 0.58ng/mL had troponin T levels of 290.0 (8.5-9510.0), while those with a value > 0.58ng/mL had troponin T levels of 132.15 (9.4-5190.0) (p < 0.012). The receiver operating characteristics (ROC) curve of apelin-36 was used to plot the true positive rate against the false positive rate at different cut-off points, with AUC=0.67 (95% CI, 0.57-0.76), and the cut-off value for apelin-36 was 0.58ng/mL, with p=0.001. Conclusions: Significant associations were observed between apelin-36 and no-reflow phenomenon in patients with STEMI. An apelin-36 cut-off value of 0.58ng/mL, measured at admission, could be used to identify patients who were at increased risk of no-reflow phenomenon/reperfusion injury.

Abstract:

Anxiety and panic attacks are among the most prevalent mental health challenges today, significantly impacting individuals’ lives, emotional stability, and overall well-being. Despite the availability of effective therapeutic techniques many individuals struggle to apply these tools consistently, particularly during acute episodes. This gap reveals the need for accessible, personalized, and engaging digital interventions that support both prevention and crisis management. This article presents the design, development, and evaluation of a digital solution that leverages Virtual Reality (VR) to assist individuals in managing anxiety. To maximize user engagement, the solution incorporates gamification elements grounded in psychological principles. The prototype was evaluated through usability testing and qualitative feedback from both mental health experts and end-users. The results confirmed the high usability and therapeutic potential of the approach as participants reported increased feelings of calmness and being better able to cope with anxiety issues.

Abstract: Opinion mining is the process of analyzing the content people create, such as product reviews or social media posts, to determine if the feelings expressed are positive, negative, or neutral. Twitter, one of the most popular social platforms for sharing opinions, provides a lot of data that can be used to understand public sentiment. In this project, we developed a system for classifying sentiments that begins with a detailed preprocessing step using natural language processing techniques. After the data has been processed, the tweets are represented using the traditional Term Frequency-Inverse Document Frequency (TF-IDF) model to highlight the most important text features.To make these features even more relevant, we introduced the Egret Swarm Optimization Algorithm (ESOA), a method for selecting important features inspired by how Great and Snowy Egrets hunt. ESOA uses three strategies—waiting patiently, actively searching, and making decisions based on differences—to find a good balance between exploring new areas and focusing on known ones. This creates a flexible framework that works well in different situations. For sentiment classification, we use a Multi-Head Attention Mechanism (MHAM) that can understand various meanings in user text. We fine-tuned the model’s settings using the Dwarf Mongoose Optimization (DMO) algorithm, along with a strategy that helps each part of the attention mechanism focus on different aspects of the text. Testing our approach on the Sentiment140 dataset shows it works very well, achieving almost 97% accuracy, which is better than other methods that usually reach between 92% and 95%.