Exploring the Role of Educational Technology in Enhancing Self-Regulated Learning Behavior Among University Students in South Korea
DOI:https://doi.org/10.65613/741121 Prof Lee Chonho Konkuk University Seoul South Korea Email: Lee.Chonho@konkun.kr Abstract：Education Technology Learning behavior has emerged as a core data-driven approach for monitoring, evaluating, and optimizing teaching and learning processes in modern higher education. Traditional student behavior prediction models rely heavily on black-box machine learning algorithms, which achieve high predictive accuracy but suffer from insufficient transparency, poor interpretability, and limited practical guiding value for educational decision-making. To address the inherent defects of opaque predictive models, this study introduces explainable artificial intelligence (XAI) technology to optimize student behavior prediction models for higher education learning analytics. This paper constructs a systematic XAI optimization framework for student behavior prediction, selects multi-dimensional student behavioral feature indicators, adopts hybrid machine learning models to complete predictive training, and utilizes SHAP and LIME interpretable algorithms to realize global and local interpretation of model decision-making mechanisms. Through empirical data verification of university online and offline learning scenarios, the optimized model not only maintains stable predictive accuracy for student academic performance, learning engagement, and dropout risk but also clearly reveals the contribution weight and action mechanism of different behavioral features. The research results effectively solve the black-box dilemma of traditional predictive models, improve the credibility and practicability of learning analytics results, and provide scientific decision-making support for personalized teaching intervention, educational management optimization, and student self-regulated learning in higher education. Meanwhile, this study summarizes the application challenges of XAI in educational predictive modeling and puts forward targeted optimization strategies, laying a theoretical and practical foundation for the intelligent and transparent development of higher education learning analytics. Key Words: learning behaviour, educational technology, effectiveness, and motivation, studies on Asia/China 1. Introduction 1.1 Progress of Learning Analytics in Higher Education With the comprehensive integration of digital information technology and higher education, smart teaching platforms, online learning systems, and campus management platforms have accumulated massive multi-source heterogeneous student learning behavioral data[1]. These data cover classroom attendance, online learning duration, resource browsing frequency, assignment completion quality, interactive discussion participation, and academic test performance, fully reflecting the whole-process learning state and behavioral characteristics of college students. Learning analytics, as a data-driven educational research paradigm, excavates implicit learning rules and behavioral patterns from massive educational data through statistical analysis, machine learning, and data mining technologies, so as to realize dynamic monitoring, early warning prediction, and personalized guidance of student learning status. In recent years, learning analytics has been widely applied in higher education teaching management, student training, and educational decision-making. It has achieved remarkable results in early warning of academic failure, identification of at-risk students, evaluation of teaching effectiveness, and optimization of curriculum setting[2]. Traditional learning analytics research focuses on improving the predictive performance of models, and continuously optimizes algorithm structures to enhance the accuracy of predicting student academic outcomes and behavioral trends. However, most mainstream predictive models such as deep neural networks, random forests, and gradient boosting algorithms belong to typical black-box models. Although these models can output high-precision prediction results, they cannot clearly explain the internal decision-making logic, the influence degree of each feature, and the causal relationship between student behaviors and learning outcomes, which greatly restricts the practical application value of learning analytics in actual higher education teaching scenarios[3]. 1.2 Defects of Traditional Student Behavior Prediction Models Traditional student behavior prediction models in higher education learning analytics have three prominent inherent defects in practical application. First, the transparency of model decision-making is insufficient. Black-box algorithms realize the mapping from behavioral features to prediction results through complex internal parameter iteration, and the decision-making process is completely concealed. Educators can only obtain final prediction results such as excellent academic performance or dropout risk, but cannot clarify which student behaviors lead to such results, resulting in inability to implement targeted teaching intervention. Second, the generalization and interpretability of models are conflicting. Most models that pursue high predictive accuracy sacrifice interpretability, while simple interpretable models often have low prediction precision and poor adaptability to complex educational data. Third, the practical guiding value of model results is limited. Traditional prediction models only complete result prediction, lacking in-depth analysis of feature contribution and behavioral mechanism. The analytical results can only serve as passive data reference, but cannot provide actionable optimization suggestions for teachers’ teaching adjustment, managers’ educational management, and students’ self-learning improvement. In addition, most traditional models adopt overall average analysis for student groups, ignoring individual differences in student learning behaviors, and cannot realize personalized interpretation and precise prediction for individual students, which fails to meet the personalized training needs of modern higher education[4]. 1.3 Value of XAI Optimization for Educational Predictive Modeling Explainable artificial intelligence, as a new branch of artificial intelligence technology, focuses on breaking the black-box barrier of traditional machine learning models, realizing transparent, credible, and understandable model decision-making. Different from traditional artificial intelligence that only pursues predictive accuracy, XAI takes model interpretability, transparency, and robustness as core optimization objectives, which can effectively make up for the defects of traditional student behavior prediction models in higher education learning analytics. The application value of XAI optimization is mainly reflected in three dimensions. At the educational management level, XAI can clarify the key influencing factors of student learning behaviors and academic outcomes, help educational managers accurately grasp the core rules of student learning development, and formulate scientific and reasonable teaching management systems and curriculum optimization schemes. At the teaching practice level, XAI can realize fine-grained interpretation of individual student behavioral characteristics, help teachers accurately identify students’ learning difficulties and behavioral deficiencies, and implement personalized teaching intervention and targeted tutoring[5-7]. At the student development level, XAI can feed back the influence of different learning behaviors on academic performance to students, guide students to recognize their own learning deficiencies, and optimize self-regulated learning strategies and learning behavior habits. In addition, XAI optimization can improve the credibility and acceptability of learning analytics results. Educational decision-making is closely related to student development and teaching quality, and opaque black-box model prediction results are difficult to be recognized and applied by educators and students. The interpretable results formed by
Algorithmic Management Dilemma: The Double-Edged Sword Effect on Employee Well-Being and Task Performance
School of Economics and Management，Sichuan Agricultural University, Mianyang, 621000, Sichuan Province, China.v455921738@163.com AbstractAn algorithmic management (AM) has emerged in the rapidly changing arena of digitalworkplaces, representing a transformative — and controversial — concept. Using a survey with312 respondents from a variety of industries where work is allocated, monitored and evaluatedusing algorithmic systems, this study examines the double edge sword of AM impact onemployee well-being and task performance. The study shows that while AM significantlyincreases task performance (measured by higher efficiency and consistency; better real timefeedback), it significantly but negatively affects employee well being (higher psychologicalstress, perceived surveillance and autonomy loss). The mediating role of perceived autonomy inthe relationship between AM and well-being is further confirmed by the mediation analysis andit reveals its crucial role in human systems interaction. This is a paradox: algorithmic tools thatare built for performance optimization can cause emotional exhaustion and alienation whenimplemented without human-centered defenses. This study enlarges the ongoing discussion ofdigital labor, AI machine work systems and workers’ resiliency and provides practical insightsfor ethical design and regulation of AM technology. The need is for an ‘and’: the harmony ofautomation and empathy to guarantee that the primal essence of human experience in work is notdigested by the efficiency of technology. Keywords: Algorithmic Management, Employee Well-Being, Task Performance, Autonomy,Digital Labor, Surveillance, AI in HR, Ethical Design, Workplace Automation, Technostress al., 2021). For example, in warehouse settings algorithmic scheduling has reduced idle time andimproved throughput (Rosenblat & Stark, 2016). However, while the same systems mightchampion metrics and micro surveillance and continuous assessment they may threatencreativity, adaptability and emotional resilience (Veen et al., 2020). Particularly sharp are gigworker reports of feeling ‘trapped in metrics’, i.e., in the sense that all their actions are undersurveillance and measured against opaque performance benchmarks (Moore, Upchurch, &Whittaker, 2018). The psychological consequences of these are consistent with the technostressframework (Tarafdar et al., 2019) that frames digital technologies as both an enabler as well as astressor based on the extent of control exercised by the user, training and support systemsassociated with these technologies.Concerns have also been raised in regards to the dehumanizing components of AM by scholars.Although traditional supervisors may be empathetic, judicious and make contextual judgments,algorithmic managers inflexibly and lacking emotional nuance impose rules (ParentRocheleau &Parker, 2021). Feedback and decision making without interaction with people erodes relationaltrust leading to feelings of alienation or depersonalized work (Kellogg et al., 2020). Additionally,employees are frequently left out of the design and iterative process for these systems and, as aresult, are left powerless to participate in and bargain collectively (Galière, 2022). It adds to theethics of digital labor rights, algorithm accountability and workplace fairness particularly insettings in which the workforce is already precarious.As such, AM is a dual technical and socio-technical phenomenon and the discussion regardingits implications for STS theory resonates with broader debates regarding the nature oftechnological interventions and their direct instrumental and indirect socio-humanistic impacts(Trist & Bamforth, 1951). Rather, AM tools are not neutral, but rather form and are formed bythe organizational environment (i.e., culture, labour relations and institutional frameworks) inwhich they are situated (Zuboff, 2019). An alternative lens to pursue this inquiry is similarlyprovided by the Job Demands-Resources (JD-R) model. According to Bakker and Demerouti(2007), job demands, e.g., surveillance and inflexible metrics, deplete energy and psychologicalresources; conversely job resources, like feedback and autonomy, enhance engagement and wellbeing. Employee strain ensues and work satisfaction declines as the number of personal resources and work demands rises, but algorithmic management results in poor resourceallocation.These tensions need to be addressed and thus, we need to empirically gauge the effects ofalgorithmic management on task performance and employee well being. Despite the surge of asignificant body of work that has focused either on performance gains or stress outcomes andwithin different employment contexts, a paucity of studies considers both performance gains andstress outcomes together. In order to address this gap, this study combines quantitativeperformance metrics with qualitative insights on worker experiences to analyze the double-edgedsword effect of AM. The aim of doing so is to take part in the development of algorithmicgovernance frameworks that are balanced and ethical, while enabling productivity withoutcompromising human dignity of labor. analytics make workers more anxious, outraged and less satisfied with their jobs when they arenot allowed to understand or influence their opaque metrics. The result of this produces whatO’Neill (2016) calls ‘weapons of math destruction’, utilizing flawed or biased data to facilitateunfair outcomes that are at a high level hard to refute.Furthermore, literature has demonstrated that algorithmic control increases work intensificationand emotional labor to unsustainable degrees. According to Fleming (2017), algorithmicmanagement replicates a kind of ‘digital Taylorism’: the minutia of an employee’s workbecomes monitored and quantified and overseen in real time. In this condition, workers areexpected (although unwillingly) to carry out the performative labor, described by Diefenbachand Sillince (2011), as working tirelessly to optimise ‘visibility’ and ‘productivity’, regardless ofthe loss of authenticity and health. Leighton (2019) documents, for example, how Amazonwarehouse employees were pushed by algorithmic weights to forgo bathroom breaks, todownplay injuries and to stay in an escalated state of productivity at the expense of burnout andattrition.On the other hand, algorithmic systems also provide opportunities to reduce bias and standardize.In one vein of research, AM tools can decrease favoritism and discrimination by ensuringuniformity by utilizing well designed AM tools to enforce uniform rules and performancestandards (Binns, 2019). For example, algorithms used in HR analytics can increase diversity byreducing unconscious bias during candidate, evaluation and resume screening, according to Kim(2017). But even these benefits rely on knowing that these systems are trained with data that istransparent and fair. However, if the training data contain past inequalities or discriminatoryhiring patterns, the algorithm will replicate and perfect these patterns, as reported by Noble(2018) in his study on algorithms bias in digital platforms.The literature in the domain of employee autonomy takes a nuanced view. However, otherscholars contend that, far from diminishing agency, AM forces conformity to rigid behavioralscripts (Scholz, 2017); and yet still other scholars argue that algorithmic tools can enhanceautonomy in specific settings. Boudreau and Lakhani (2013)
A Cross-Center Risk Prediction Model for OsteoporoticFracture Under the Federated Learning Framework
First author: Yizhe Fan, drfanyizhe@163.comSecond author and co-first author: Zhongyuan Shen, zyshen@njucm.edu.cnThird author: Xiao Zhang, qucyzhang@163.comFourth Author: Zhen Han, 18168988971@163.comCorresponding author: Chengjian Wei, drweichengjiantcm@163.comAcknowledgementFundingsThe current work was supported by the National Nature Science Foundation ofChina (No.81973872); Jiangsu Provincial Medical Key Discipline (Laboratory)Cultivation Unit (No. JSDW202252); Postgraduate Research & Practice InnovationProgram of Jiangsu Province (No.SJCX24_0960).Ethics approval and consent to participateThis study was established and authorized by the Animal Care and UseCommittee of the Nanjing University of Chinese Medicine (Approval number:ACU231001). AbstractThis work sets out to build a privacy-preserving risk-evaluation engine forosteoporotic fractures, stitched together across several clinical sites rather than parkedin one central warehouse. To do so, the authors lean on a federated-learning scaffoldthat lets participating hospitals crunch their own numbers, ship only gradients back toa neutral server, and still compare performance in a meaningful way without tradingpatient identities. Researchers lifted the patient sample from NHANES 2017-2020,slicing it into three virtual centres that mimic the demographic and technicalpatchwork seen in community care; each slice held 2,743 individuals in total andcarried 42 columns describing things like age, lab values, lifestyle habits, and bonemineral density. The base learner is a compact neural net dressed with differentialprivacy, Byzantine guards, and gradient-compression tricks so the server side staysmanageable even under heavy load. Adaptive weighting schemes soften the usualaches caused by uneven data distributions, letting the architecture dodge the poorgeneralisability that plagues most single-site prototypes. Final numbers read 0.847 forarea under the curve, a shade ahead of the comparable centralised version (0.832, p =0.024) and miles better than the rough-and-ready FRAX calculator (0.734, p 0.001).Even when noise budgets tighten to ε = 1.0, leakage metrics drop by 60 and the AUCstill rests at a respectable 0.841. A post hoc consistency check shows centre-by-centrescores clustering tightly between 0.841 and 0.853, reassuring the research team thatthe framework scales well no matter whose data plug into the engine.Recentinvestigations have shown that federated learning sidesteps the privacy bottlenecksthat typically beset single-centre trials. By distributing the analytics across multiple 2 nodes rather than hauling patient records to a central vault, the approach keepssensitive information resident at its source. Researchers now regard that model as thefirst truly scalable blueprint for multi-institutional collaborations in clinical machinelearning.Keywords: Federated Learning; Osteoporotic Fracture; Risk Prediction;Differential Privacy; Machine Learning; Multi-center Collaboration 3 estimating fracture risk—rather underexplored [12] . Typical federated learningdeployments face persistent hurdles, including the stark data heterogeneity foundacross different contributing sites, the pressing need for efficient communicationprotocols, and the constant demand for privacy safeguards that match the sensitivityof medical information [13] . Merging federated algorithms with routine clinical riskworkflows also compels practitioners to reckon with dual requirements: theinterpretability of the distributed model outputs and the rigorous clinical validationthat decision-makers expect [14] .This project confronts well-documented barriers in osteoporosis research byconstructing a wide-reaching federated-learning infrastructure tailored to predictingfracture risk across separated hospital networks. Embedded within the framework arecutting-edge, privacy-minding routines, sophisticated aggregation schemes thatreconcile uneven data distributions, and interpretable algorithms that clinicians canreadily discuss with patients. Field tests draw on openly released datasets that mimicmulti-centre patient flows, offering early proof that distributed techniques can rivalconventional single-site models in clinical forecasting for bone health. By keepingraw records stationary and exchanging only numerical summaries, the methodologysafeguards personal information while still permitting large-scale learning. Outputsfrom the system not only advance the technical literature on federated health AI butalso lay groundwork for tomorrow’s collaborative decision-support tools in hospitalsettings. In practical terms, the infrastructure equips different care providers to poolinsights without crossing the regulatory red lines around data movement, thusbroadening the evidence base for everyday fracture prevention strategies. 4 standard centralised methods and common clinical scorecards; results were filed insuch a way that anyone with the public datasets could repeat the exercise.2.2 Feature Engineering and Data PreprocessingA detailed data-preprocessing pipeline was built to safeguard quality anduniformity among the various simulated centres, as shown in Figure 1. Within thatframework, engineers carried out feature extraction, filled in missing values, flaggedoutliers, and standardised the data—each step vital to the reliable training of machine-learning models. NHANES Raw Data2017-2020 Cyclesn= 15,560 subjectsFeature ExtractionDemographicsAge, Gender, BMlBMD T-scoresSpine, Hip, NeckLaboratoryCa, P, Vit DData Quality ControlMissing DataKNN lmputationk=5 neighborsMissing rate < 20%Outlier DetectionlQR MethodQ1-1.5xIQR toQ3+1.5xIQRData ValidationRange ChecksClinical rangesConsistencyFeature StandardizationZ-score NormalizationZ=(X-μ)/σPer-center basisFeature EncodingOne-hot EncodingNominal variablesRace, medication useOrdinal EncodingOrdered categoriesEducation levelProcessed Dataset89 features, Ready for FL Figure 1: Data Preprocessing and Feature Engineering Pipeline Feature extraction in the present study set out to delineate those variables moststrongly linked to the hazard of osteoporotic fracture. Basic demographicmarkers—age, sex, racial background, and body mass index—enter the analysis as afoundation. Clinical inputs then follow, notably the lumbar spine and total hip Tscores plus femoral neck T values, together with serum calcium, phosphorus, and 25-hydroxyvitamin D levels. Lifestyle contributors rely on self-reported smoking,habitual drinking, activity frequency, and daily calcium intake from food orsupplements. A final block captures patient history, listing prior fractures, a familyosteoporosis pedigree, and any medications currently in use.Missing data imputation was performed using k-nearest neighbors (KNN)algorithm to preserve the underlying data distribution. For continuous variables, theimputed value was calculated as: 11kiiimissingkiiwX X w where iwrepresents the inverse distance weight for the i -th nearest neighbor 5 and iXdenotes the corresponding feature value.Outlier detection employed the interquartile range method, identifyingobservations beyond 31.5QIQR or below 11.5QIQR , where 1Qand 3Qrepresent the first and third quartiles, respectively. Extreme outliers were winsorizedto the 95th or 5th percentiles to maintain data integrity while preserving sample size.Feature standardization was implemented using Z-score normalization to ensurecomparable scales across different measurement units: XZ where and represent the population mean and standard deviation,respectively. This standardization process was performed independently within eachsimulated center to preserve the federated learning paradigm and prevent data leakagebetween institutions.Categorical variables were encoded using one-hot encoding for nominalvariables and ordinal encoding for naturally ordered categories. Feature correlationanalysis was conducted to identify highly correlated variables (|r| > 0.8) andimplement appropriate dimensionality reduction strategies when necessary.2.3 Federated Learning Framework DesignA federated learning architecture was created to facilitate joint prediction ofosteoporotic-fracture risk among a network of virtual healthcare providers, all withoutexposing individual patient records. In building the framework, the researchersclosely
Research on the Enhancement Mechanism of Rural Teachers’ Professional Attractiveness in the Context
First author: Yu Zhang, 15246339088@163.comSecond author and corresponding author: Ying Zhen, zhenying0322@163.comThird author: Chen Qing, 13978848156@126.com AcknowledgementNational Social Science Fund of China 2021 Education General Project “Research on RuralTeacher Retention Mechanism” (No. BHA210137)Youth Project of Philosophy and Social Sciences Research Plan in Heilongjiang Province“Research on the Integration of College Teacher Training and Rural Basic Education Needsunder the Background of Rural Revitalization Strategy” (No. 21EDC199) AbstractThis study is a response to the persistent challenges of rural teacher recruitment and retention,exploring a multi-dimensional enhancement mechanism for the professional attractiveness ofrural teachers within the framework of China’s rural revitalization strategy. The study uses a mixed methods approach to identify institutional support, psychological motivation, financialincentives and community engagement as the most influential predictors of professionalattractiveness using quantitative survey data from 428 rural teachers of varying age, experienceand qualification levels from five provinces, supported by factor and regression analysis.However, the findings indicate that economic benefits are not sufficient and must beaccompanied by career development pathways, inclusive school governance, as well asemotional ‘fulfillment’ for a long term commitment. The subsequent analysis, on a regional basis,shows that teachers’ perceptions of their profession are mostly conditioned by localized policyimplementation. Results are valuable for policymakers and education administrators as theproposed model for enhancement is an integrated one comprising structural and humanisticreforms. The study adds to educational equity literature and strengthens the case for all actors insystemic policy to realign policy to address teacher shortages in underserved regions.KeywordsRural education, teacher retention, professional attractiveness, rural revitalization, institutionalsupport, psychological motivation, educational policy, China “Professional attractiveness” is constituted by tangible and intangible elements including salary,opportunities for occupational progression and access to training, as well as intrinsic jobsatisfaction, social recognition and integration into the community (OECD, 2018; Guarino et at.,2006). For example, Rakenuy (2022) shows that studies have long shown that job security,professional development supports and a mission motivate teachers to work in rural schools(Ingersoll, 2017; Zhao, 2021). However, rural teaching positions have been called ‘career deadends’ because of little route for advancement, under resourced settings and lack of administrativesupport (Xu & Hu, 2020). Additionally, differences in living standards, schools and educationalinfrastructure and social status between rural and urban teaching settings have exacerbated thisview which in turn contributed to high turnover and long absence from teaching (Chen & Sun,2019).China’s rural revitalization campaign has been launched in 2018 and rural teaching has onceagain regained its popularity as a viable and rewarding career (State Council of the People’sRepublic of China, 2018). The strategy is dedicated to a holistic improvement of rural regions onthe themes of education, health care, the economy and culture. However, without targetedmechanisms that increase the attraction and sustainability of rural teaching roles such reformsmay not realize their full potential (Zhang & Huang, 2020). Policy efforts, including the “SpecialPost Program”, provide financial and job security incentives to recruit new graduates to ruralposts (MoE China, 2020), but their long term effectiveness remains unconfirmed based on theabsence of the systematic support structure and the localized implementation strategy (Wang &Yang, 2022).And indeed, global research shows the same. Teacher attrition in remote areas in low and middleincome countries is also driven by things besides pay: professional isolation, poor housing, lackof peer support and few family amenities according to the World Bank (2021). Indeed, UNESCO(2020) has also emphasized the need for localized policy making which respects diversity interms of cultural and regional diversity to retain teachers for rural education. This backdropraises a critical research need, namely to understand the mechanisms that have the potential totruly augment rural teachers’ professional attractiveness in countries undergoing rapid ruralrestructuring. Addressing this gap this study examines multifaceted determinants of rural teachers’ professionalappeal within the process of rural revitalization. This paper presents an actionable frameworkthat chronicles how economic, institutional, psychological and sociocultural factors operatetogether to influence rural teachers’ decision regarding career choices through a synthesis ofpolicy evaluation, empirical evidence and teacher perceptions. Specifically, the findings aim tocontribute to academic literature and policy design that could inform the development of adurable, egalitarian and high quality rural education ecosystem. communities. Per Chong and Low (2009), teachers, in particular in isolated teachingenvironments, can work out reasons that make this a proper profession to educate in. However,teacher training institutes continue to be both curriculum and field experience urban centered,thus creating a misalignment of training and deployment needs (Reid et al., 2010). Additionally,Avalos (2011) points to the lack of continuous learning opportunities in rural contexts as thereason for the diminished professional appeal that rural teachers experience because the absenceof workshops, certifications or access to higher education makes rural contexts feel stagnant forteachers.A second major area addressed in global research is related to the social and cultural integrationof the rural teachers in the community. Isolation is still seen by Hargreaves (2001) as the maincause of attrition whilst posting rural teachers. Miller (2012) found that teachers with strongcommunity engagement in rural Australia were more likely to remain in their profession forlonger periods of time. Recognition from the community and from the society at large lends asense of belonging, much of which is very critical especially in environments where there is lowinstitutional support. Teachers’ social capital and professional self worth can be improved bycommunity based recognition schemes, local housing integration and participation in culturalactivities (Sharplin, 2008).Recent studies have begun to investigate rural revitalization policies in the intersection witheducational reforms in the Chinese context, for instance what leads to the attractiveness of ruralteaching. Local governments are beginning to test the “educational ecology” models—holisticeducation that overlaps school development with community revitalization (Li and Yang 2018).The ultimate goal of these initiatives is to do more than just modernize facilities, but develop anambiance meant for a satisfying teacher life. There is, however, an unevenness toimplementation and the distribution of educational investments has an urban bias (Zhou, 2020).Moreover, there is a growing interest in psychological and emotional well being as a centralelement in attracting and retaining teachers. Skaalvik and Skaalvik (2011) in a longitudinal studyfound emotional exhaustion, perceived lack of support and workload pressure to be significantpredictors for teachers leaving their profession and rural teachers more specifically
The influence of leisure sports recreation resources value on recreationers’ willingness to revisit
1.Male, school of Leisure sprots,Chengdu Sport University,Chengdu City610041,Sichuan, China2.Female,College of physical education and health ,Aba Teachers University, AbaTibetan and qiang autonomous prefecture 623002,Sichuana.Email：101054@cdsu.edu.cnb.Email：20139613@abtu.edu.cn Abstract: Leisure sports recreation resources serve as the foundation forrecreational activities, encompassing diverse categories such as natural andcultural resources. Their value is constituted by functional, experiential,emotional, and social dimensions. Revisit intention, a key indicator of tourismexperience and destination appeal, is influenced by factors like satisfactionand emotional identification. This study analyzes the components of leisuresports recreation resource value and the determinants of revisit intention,revealing that resource value exerts influence through a sequential pathway:“value perception → satisfaction enhancement → emotional reinforcement →revisit intention formation.” Based on these findings, we propose strategiesfor enhancing resource value, improving revisit intention, and achievingsynergistic optimization between the two aspects, providing actionablereferences for the sustainable development of leisure sports recreationdestinations.Key words: leisure sports recreation resources; revisit intention;influence analysis; optimization strategyforewordWith the steady implementation of China’s “Healthy China” strategy, leisuresports tourism has become an integral part of modern life. The development andsustainable management of these resources have garnered significant attention.As a key indicator of destination competitiveness, repeat visit willingnessplays a vital role in maintaining tourist retention and reducing marketingcosts. While current leisure sports tourism resources exhibit diversifiedcharacteristics, their intrinsic connection between resource value and visitorloyalty remains underexplored. Clarifying this relationship is crucial forenhancing resource appeal and driving high-quality industrial development.I. Definition and classification of leisure sports recreation resourcesLeisure sports recreation resources refer to the comprehensive collection of material and intangible elements that meet people’s needs for leisure,fitness, and entertainment. These resources form the foundational conditionsfor conducting recreational sports activities. They can be categorized based onresource attributes and functional characteristics: In terms of resourceattributes, they include natural categories such as mountains, lakes, andforests suitable for outdoor sports, as well as cultural categories like sportsvenues, fitness trails, and sports theme parks[1]. Functionally, there arefitness and leisure facilities (e.g., gyms, square dance areas), competitiveviewing venues (e.g., sports event stadiums), adventure-oriented resources(e.g., rock climbing bases, rafting rivers), and folk sports venues (e.g.,traditional martial arts training grounds, dragon boat racing waters).Together, these diverse resources establish a multifaceted leisure sportsrecreation system.Subcategorization Specific examples natural kind Mountains, lakes, forests and other natural Spacessuitable for outdoor sports Humanities Sports venues, fitness trails, sports theme parksand other artificial facilities Fitness andrecreation Gyms, square dancing venues Competitiveviewing A venue for a sporting event Outdooradventure Rock climbing base, river for rafting Folk sportscategory Traditional martial arts practice ground, dragonboat race water area Table 1 Classification of leisure sports recreation resourcesThe concept of willingness to revisitRevisit intention refers to the subjective inclination and psychologicalwillingness of tourists to return to a specific destination after completing a travel experience. As a key indicator for evaluating tourism quality anddestination appeal, it is influenced by multiple factors including touristsatisfaction, emotional identification, destination image, and service quality.Common manifestations include clear plans for future visits, willingness torecommend the destination to others, or sustained interest in the location[2].Revisit intention not only reflects tourists ‘recognition of their travelexperiences but also correlates with sustainable development of destinations.High revisit intention helps reduce marketing costs, stabilize tourist markets,and serves as a core element demonstrating a destination’s competitiveness. felt during traditional dragon boat races—strengthens their preference forreturn visits. Resource uniqueness is equally crucial: scarce naturallandscapes (e.g., exclusive hiking trails) or distinctive activities (such asintangible cultural heritage sports experiences) can reduce alternativedestination appeal[3]. Word-of-mouth recommendations from friends and family,social media reviews, and destination innovation (like new activity offeringsand seasonal events) all influence participants’ psychological expectations anddecision-making, ultimately shaping their willingness to revisit.(3) The relationship between the value of leisure sports recreation resources and thewillingness to revisitThe value of leisure sports recreation resources serves as a key driver forrepeat visitation intentions, demonstrating a strong positive correlationbetween the two. The functional value of resources directly enhances basicsatisfaction by aligning with recreational needs and practical expectations,thereby laying the foundation for repeat visits. For instance, well-equippedprofessional fitness facilities encourage regular return visits from fitnessenthusiasts. Experiential and emotional values strengthen psychologicalconnections to boost revisit loyalty. Immersive experiences like forest cyclingor emotional resonance during team-building activities create lasting nostalgiaand revisit impulses. Additionally, the social value and uniqueness ofresources reinforce destination irreplaceability, reducing visitor switching.When visitors perceive resource value as closely matching their needs, theydevelop trust and reliance on the destination, transforming resource valueperception into explicit revisit tendencies. This establishes a transmissionpath: “value perception → satisfaction enhancement → emotional reinforcement→ repeat visit intention formation”. cultivation requires cultural empowerment by exploring regional sportstraditions, transforming folk activities like dragon dance and polo intointeractive experiences, and enhancing emotional connections through narrativestorytelling. For social value expansion, establishing “sports + social”platforms is crucial. Regular community sports events and family-friendlycarnivals should be organized, while resource-sharing mechanisms like schoolsports facilities being open to the public during holidays should beimplemented. This creates a closed-loop value enhancement system characterizedby “comprehensive functionality, unique experiences, emotional bonds, andsocial recognition” [4].(2) Strategies to enhance the willingness of recreationers to revisitTo enhance visitors ‘willingness to return, it’s essential to establish amulti-touchpoint incentive system spanning the entire lifecycle. Experienceoptimization forms the foundation of this framework, covering every stage ofrecreation: Before travel, VR technology provides virtual destination tours tohelp tourists plan their itineraries. During activities, a “tailored service”mechanism is implemented – offering childcare for families and professionalcoaching for fitness enthusiasts. Post-experience, personalized reports withexercise statistics and memorable photos are emailed. For emotional retention,long-term interaction mechanisms should be built: First, create exclusiveprofiles documenting participants’ exercise preferences and activity records.Then, send customized birthday wishes and event invitations during holidays tofoster a “sports community” that enhances belonging. Organize online groups bysports type and regularly host offline events like monthly themed running clubsto build stable interest communities. Differentiated attraction is key toincreasing return visits. Develop unique experiences through resource-specificinitiatives: Lake-based projects like “Starry Night Kayaking Tours,” ormountain-themed challenges like “Seasonal Hiking Challenges” – these rareexperiences reduce the likelihood of being replaced. To enhance engagement, weimplement a tiered reward system featuring a “Revisit Points System” whereaccumulated points can be redeemed for sports gear, exclusive courses, oraccommodation discounts. High-frequency visitors receive “Honorary Membership”privileges including free parking and priority
A study Of Argentina Philosophy and culture
CARLOS GAMERRO DOI:http://doi.org/10.65613/742454 Professor, Argentina university of Arts Email: ：carlos.gamerro@gmail.com Abstract: Argentina With the development of fintech and popularization of wealth management concepts, intelligent investment consultants have become important in the wealth management market, offering low – cost, efficient and personalized services. Traditional strategies rely on the “rational man” assumption, ignoring investors’ behavioral biases like loss aversion. Loss aversion, proposed in prospect theory, means investors feel more pain from losses than pleasure from equivalent gains. This bias leads to irrational decisions, affecting investment returns and service effectiveness. Therefore, exploring the optimization path of intelligent investment consultant strategies from the perspective of loss aversion is significant. This paper first combs related theories, analyzes the impact of loss aversion on decision – making and the defects of traditional strategies. Then, based on loss – aversion characteristics, it proposes targeted optimization schemes in investor portrait construction, asset allocation, risk control and investor education. Finally, it verifies the optimized strategy’s effectiveness through empirical analysis and gives relevant suggestions. The research shows that the optimized strategy can better adapt to investors’ behavior, reduce the negative impact of loss aversion, improve investment experience and returns, and enhance service competitiveness. Keywords: Loss Aversion; Intelligent Investment Consultant; Strategy Optimization; Prospect Theory; Asset Allocation 1. Introduction In recent years, driven by AI, big data, and cloud computing, the fintech industry has developed rapidly, and the wealth management market has entered a digital and intelligent era. Intelligent investment consultants, a new wealth management service model, integrate financial theory, algorithm models, and technology. They can generate personalized investment portfolios based on investors’ risk preferences, goals, and asset status, and adjust them with market changes[1]. Compared with traditional manual investment consultants, they have advantages like low threshold, low cost, high efficiency, and strong objectivity, and are widely recognized by investors. By 2025, China’s intelligent investment consultants’ asset management scale exceeded 500 billion yuan, and the number of users exceeded 100 million, showing strong momentum. However, as the market scale of intelligent investment consultants expands, problems have emerged. Traditional strategies are often based on classical financial theories assuming investors are rational. But in reality, investors are affected by behavioral biases, especially loss aversion. Kahneman and Tversky (1979) proposed the prospect theory, stating that loss aversion is common. The psychological utility loss from a loss is 2 – 2.5 times the utility gain from an equal gain. This makes investors over – emphasize loss risks and make irrational decisions, such as holding loss – making assets or selling profitable assets prematurely[2-4]. These behaviors reduce investors’ returns and the effectiveness of recommended portfolios, decreasing user trust and satisfaction. Currently, most research on intelligent investment consultants focuses on algorithm optimization, asset allocation efficiency, and service scope expansion. Few studies combine loss aversion for strategy optimization[5]. The mismatch between traditional strategies and investors’ behavioral characteristics (especially loss aversion) restricts the industry’s development. Therefore, it is necessary to study the impact of loss aversion on decision – making, find defects in traditional strategies, and propose targeted optimization schemes. This study starts from loss aversion to explore the strategy optimization path, which can enrich theories and guide the industry’s development, meet investors’ needs, and promote the healthy development of the wealth management market. 2. Theoretical Basis 2.1 Loss Aversion Theory Loss aversion, the core of prospect theory proposed by Kahneman and Tversky in 1979, is the psychological tendency where investors feel far more pain from losses than pleasure from equivalent gains. In prospect theory, they introduced the value function to describe the relationship between investors’ subjective value and gains/losses. The value function has three features: it uses a reference point as the origin, with gains and losses relative to it; it’s convex in the loss area and concave in the gain area, meaning investors are risk – averse in the gain area and risk – seeking in the loss area[6]; its slope is steeper in the loss area, showing losses have a greater impact on psychological utility. For example, losing 100 yuan causes much more pain than gaining 100 yuan, and investors usually need to gain 200 – 250 yuan to offset the pain of losing 100 yuan. Loss aversion is a common psychological phenomenon present in various investment scenarios, like in the stock market where investors often hold losing stocks for a long time and are unwilling to… Investors hold losing stocks to avoid “real” losses and sell profitable ones to lock in profits, resulting in “winning small and losing big”. In the fund market, investors panic and redeem when the net value falls, intensifying fluctuations, and redeem when it rises, missing appreciation opportunities. These are manifestations of loss aversion, affected by factors like psychological quality, investment experience, risk capacity, and market environment. Novice investors show more obvious loss aversion due to lack of experience and stronger fear of losses. Experienced investors can restrain it but can’t fully avoid its impact in extreme market fluctuations. Loss aversion affects investors’ decision – making in two ways: it makes them overly risk – averse, preferring low – risk, low – return assets and missing high – return opportunities; it also leads to irrational decisions like blind stop – loss and holding losses, reducing investment returns and even causing serious losses[7-9]. Therefore, understanding loss aversion’s characteristics and impact is the premise of optimizing the intelligent investment consultant strategy. 2.2 Intelligent Investment Consultant Intelligent investment consultant, or robo – advisor, is a financial service model using AI, big data, and algorithm models to collect and analyze investors’ information and offer personalized investment advice and asset allocation services. Its core is the algorithm model, including asset allocation, risk assessment, and market prediction models. The working process has four steps: first, collect investors’ information like risk preference, investment goal, etc.; second, assess investors’ risk – bearing capacity and needs via the risk assessment model; third, generate personalized investment portfolios with stocks, bonds, etc. according to the asset allocation model; fourth, adjust the portfolio dynamically based on market and investor needs to meet expected
Conjectures about the memory of my tribe: Truncated memory
DOI:http://doi.org/10.65613/742416 CECILIA GARCÍA-HUIDOBRO Ecuador Latin America University Abstract To address the real-time and robust nature of robot obstacle avoidance strategies in dynamic environments, this paper proposes a DRL-based action decision and dynamic obstacle avoidance optimization method. This approach constructs an action policy network centered on the PPO algorithm, introducing a multi-objective reward function to jointly constrain safety, path efficiency, and trajectory smoothness. State inputs fuse LiDAR and inertial navigation data to achieve spatial modeling and velocity estimation for static and dynamic obstacles.The action space employs a continuous control model, outputting linear and angular velocity commands. Simultaneously, an obstacle prediction and fallback mechanism is integrated into the policy layer to enhance obstacle avoidance success rates in highly dynamic environments. The system constructs multi-scenario simulation test environments on the ROS+Gazebo platform and conducts comparative experiments with traditional DWA and non-predictive PPO strategies.Results demonstrate that in high-density dynamic obstacle environments, the proposed method achieves a collision rate below 2.3%, elevates target arrival rate to 96.7%, reduces average path length by 12.5%, and maintains average decision latency under 112 ms, exhibiting strong generalization capability and decision stability. This research provides theoretical support and algorithmic pathways for autonomous behavior planning in mobile robots navigating dynamic complex scenarios, offering broad engineering application potential. Keywords Deep reinforcement learning; Dynamic obstacle avoidance; Robot action decision-making; State modeling; Trajectory optimization Enhancing robotic motion decision efficiency and obstacle avoidance robustness in highly dynamic environments has become a key research direction for intelligent mobile platforms.Xiao et al. [1] constructed an obstacle avoidance architecture integrating trajectory tracking and policy update mechanisms, validating deep reinforcement learning’s decision-making capabilities in continuous control scenarios. Yang et al. [2] introduced a risk-constrained optimization model to dynamically balance path efficiency and safety redundancy. Hu et al. [3] focused on path planning in unknown environments, proposing a perception-action mapping method based on deep policy iteration to enhance navigation generalization.Lyu et al. [4] optimized action output continuity by refining the DDPG algorithm, enabling robots to achieve higher trajectory stability on complex paths. Park et al. [5] embedded deep reinforcement learning within a digital twin platform, successfully deploying a sensor fusion-based action strategy system in highly enclosed nuclear power scenarios. These studies demonstrate the significant potential of deep reinforcement learning frameworks in multi-source perception fusion, action strategy generation, and complex trajectory planning. Despite progress in static environments or under regular obstacle conditions, existing methods exhibit three limitations: first, inadequate modeling capabilities for high-density dynamic disturbances, leading to strategy instability under behavioral changes; second, insufficient resolution in action output structural control, hindering adaptation to practical robotic control interfaces; third, the absence of a systematic training framework integrating perception prediction, dynamic obstacle avoidance, and trajectory optimization.To address these challenges, developing a deep policy model for continuous action spaces with trajectory prediction capabilities and multi-objective reward coordination mechanisms is crucial. By adopting a dual-network architecture based on PPO and incorporating velocity prior constraints with path smoothing penalties, the proposed approach enhances control stability while enabling feedforward adaptation to dynamic obstacle behaviors.Within the constructed ROS+Gazebo multi-scenario simulation platform, the proposed system maintains a low collision rate and optimal path efficiency even under highly complex interference scenarios, demonstrating strong deployment versatility and system engineering value. In robotic motion control research, rule-based path generation methods like DWA and APF are commonly used for near-field obstacle avoidance. However, their decision-making processes heavily rely on local information and fixed thresholds, making them ill-suited for scenarios with dynamic uncertainties. Building upon this, machine learning techniques have been progressively integrated into the action decision chain. By enhancing state representation capabilities and strategy generalization, these methods significantly improve robotic autonomy in unstructured environments [6].As shown in Table 1, while DWA offers computational efficiency advantages, its local velocity window model lacks dynamic trajectory prediction capabilities. Imitation learning can achieve human-like control strategies with sufficient training data, but its robustness significantly degrades under abnormal conditions due to the strategy’s non-updatability.In contrast, deep reinforcement learning (DRL)-based methods such as PPO and SAC construct state-action mapping functions ( ). This enables dynamic optimization of policy parameters ( ) during interactions and maximizes the value of long-term control behavior through cumulative discounted return functions. Table 1 Comparative Analysis of Action Decision Model Structures and Performance Decision Method State Modeling Approach Control Output Type Policy Update Capability Dynamic Adaptive Dynamic Weighted Average (DWA) Local velocity window-based Discrete velocity pairs × (Fixed Rules) Weak (Prone to Local Optima) Imitative Learning (BC) Expert trajectory mapping Discrete or low-dimensional continuous control × (Offline Learning) Medium (Depends on coverage) DRL (PPO/SAC) Fusion of historical states and sensor inputs High-dimensional continuous actions (v, ω) ✓ (Continuous Optimization) Strong (policy generalizability) In deep reinforcement learning-based robotic action decision systems, environment modeling not only determines the structural complexity of the state space but also directly influences the perceptual boundaries and generalization capabilities of the policy network. In this study, the environment space is constructed as a two-dimensional discrete-continuous hybrid domain containing static structural obstacles and dynamic moving entities. Here, denotes the set of immovable obstacle regions, represents the set of dynamic obstacles with velocity vectors, indicates the target point, and signifies the boundary topology[7]. To enhance the training stability of reinforcement learning strategies in complex scenarios, the velocity vectors of all dynamic obstacles in the environment are modeled as smooth trajectory functions with Poisson disturbance terms, defined as: (1) where denotes the initial velocity of the obstacle, controls the perturbation amplitude, represents the perturbation frequency, and is the Gaussian noise perturbation term, simulating trajectory deviations and non-periodic disturbances of dynamic obstacles in real environments. Figure 1 illustrates the spatial element composition within the constructed training environment, comprising multi-source obstacles, target points, and path constraint sets for dynamic moving bodies. The motion states of dynamic elements serve as critical input variables during the state modeling phase [8].This modeling approach enables the policy network to learn trajectory prediction and risk avoidance capabilities in dynamic scenarios during training, providing a necessary foundation for subsequent state perception fusion and policy optimization. Figure 1 Training