Causal Inference Techniques Research Articles

Integrating machine learning with causal inference for enhanced system dynamics modeling: A framework for predicting complex interactions

Approaches and methods like System Dynamics Modelling (SDM) have been significant for assessing the behavior of many systems. However, classical methodologies applied in traditional approaches to SDM fail to identify nonlinear feedback and power dependencies, revealing hidden temporal casual relationships. In this paper, I present a novel approach that combines ML and causal inference methods to improve the forecast capability and semantics of system dynamics models. Incorporating ML algorithms for predictions and Causal Inference techniques for explanation, this combined strategy presents a new era for understanding the system interaction and quantifying the hidden causes within various systems. We illustrate the advantages of the suggested framework over traditional SDM and purely ML approaches by employing it to analyze a genuine circumstance for both prognosis and discovering causal relationships. Our findings indicate that such integration is effective in enhancing the comprehension of system interactions and deriving a reliable method for estimating subsequent state conditions in complex contexts. The results are relevant for various disciplines, starting with economics and ending with environmental protection sciences, where interactions and changes vary. As a result, it will give a foundation for further studies of integrating future computerized methods in the dynamical system modeling of the next generation.

Application of artificial intelligence to study the causality in public health: a systematic review

Abstract Introduction In epidemiology, establishing causation requires meeting rigorous criteria beyond mere association. Traditional methodologies often struggle to provide definitive causal conclusions due to confounding variables. However, leveraging artificial Intelligence (AI) and machine Learning (ML) offers a promising solution by modeling complex interactions among relevant factors. This systematic review explores the AI techniques utilized to uncover causal relationships in public health applications. Methods A literature search was conducted in Pubmed, Web of Science, and Scopus, employing the following search strategy: (Causalit*[Title/abstract] OR Causation*[Title/abstract] OR Causal[Title/abstract]) AND (‘Machine Learning’[Title/abstract] OR “Deep Learning” [Title/abstract] OR ‘Artificial Intelligence’ [Title/abstract] OR Algorithm* [Title/abstract]). The search aimed to identify studies encompassing causal inference techniques utilizing ML and AI methodologies in public health. Results From a total of 28,230 articles published up to August 2023, after removing duplicates (11,158) and articles that did not meet the inclusion criteria (16,168 post title/abstract screening, and 856 post full-text screening), the systematic review included 48 articles. Bayesian additive regression trees, Bayesian network models, causal forests analysis, and graphical causal models emerged as the prevailing causal methodologies. AI-driven causal methodologies have been investigated across various domains within public health, encompassing disease prediction, evaluation of treatment efficacy, identification of risk factors, analysis of health behaviors, precision Public Health initiatives, and environmental health assessment. Discussion While our systematic review affirms the potential of AI-driven causal approaches across several public health domains, further research is warranted before these emerging techniques can be effectively integrated into real settings. Key messages • In epidemiology, establishing causation requires meeting rigorous criteria beyond mere association and traditional methodologies often struggle to provide definitive causal conclusions. • This systematic review explores the AI techniques utilized to uncover causal relationships in public health applications.

Automated CT-based Decoupling of the Effects of Airway Narrowing and Wall Thinning on Airway Counts in Chronic Obstructive Pulmonary Disease.

We examine pathways of airway alteration due to wall thinning, narrowing, and obliteration at different COPD severity stages using CT-derived airway metrics. Ex-smokers (N = 649; age mean±std: 69 ± 6years; 52% male) from the COPDGene Iowa cohort (September 2013-July 2017) were studied. Total airway count (TAC), peripheral TAC beyond 7th generation (TACp), and airway wall thickness (WT) were computed from chest CT scans using previously validated automated methods. Causal relationships among demographic, smoking, spirometry, COPD severity, airway counts, WT, and scanner variables were analyzed using causal inference techniques including direct acyclic graphs (DAGs) to quantitatively assess multi-pathway alterations of airways in COPD. TAC, TACp, and WT were significantly lower (p < 0.0001) in mild, moderate, and severe COPD compared to the preserved lung function group. TAC (TACp) losses attributed to narrowing and obliteration of small airways were 4.59, 13.29, and 32.58% (4.64, 17.82, and 45.51%) in mild, moderate, and severe COPD, while the losses attributed to wall thinning were 8.24, 17.01, and 22.95% (12.79, 25.66, and 33.95%) in respective groups. Different pathways of airway alteration in COPD are observed using CT-derived automated airway metrics. Wall thinning is a dominant contributor to both TAC and TACp loss in mild and moderate COPD while narrowing and obliteration of small airways is dominant in severe COPD. This automated CT-based study shows that wall thinning dominates airway alteration in mild and moderate COPD while narrowing and obliteration of small airways leads the alteration process in severe COPD.

Statistical challenges and solutions in multidisciplinary clinical research: Bridging the gap between

This paper delves into the intricate challenges and innovative solutions in applying statistical methodologies within clinical research, aiming to bridge the gap between biostatistics and medicine. The study meticulously examines fundamental biostatistical concepts, addressing the complexities of modern clinical trials and observational studies. Through a comprehensive review of advanced regression models, causal inference techniques, and machine learning algorithms, the paper illuminates the evolving landscape of biostatistics in handling high-dimensional data and confounding variables. The methods employed in this study involve an extensive analysis of current literature, case studies, and practical applications that demonstrate the utility of these advanced methodologies. Key findings reveal that traditional statistical approaches often fall short in capturing the complexities of clinical data, necessitating the adoption of more sophisticated techniques. The integration of non-linear regression models, robust causal inference methods, and machine learning has significantly enhanced the accuracy and reliability of research outcomes, offering deeper insights into patient outcomes and treatment efficacy. Conclusions drawn from this study underscore the critical need for a paradigm shift in clinical research, moving beyond the rigid reliance on p-values towards a more holistic approach that emphasizes effect sizes, confidence intervals, and practical significance. The paper recommends continued innovation in statistical methodologies, particularly the integration of big data analytics and machine learning, to address the growing complexities of biomedical data. Furthermore, it advocates for interdisciplinary collaboration and ethical considerations in the application of these advanced techniques to ensure that biostatistics continues to contribute meaningfully to the advancement of medical science.

Revisiting the Occupational Health Impact of Right-to-Work Laws: A Research Note.

This research note reevaluates the occupational health impact of right-to-work (RTW) legislation, incorporating recent developments in causal inference techniques. In an era marked by an uptick in the adoption of anti-union legislation and increases in workplace fatalities and injuries, it is particularly urgent to examine the extent to which RTW laws affect workers' health. Using a state-year-level dataset spanning 28 years and collected from multiple data sources, we apply an innovative generalized synthetic control method to overcome several limitations of the traditional two-way fixed-effects approach to examine the effect of RTW laws on occupational fatal injuries as well as various other health outcomes. Robustness checks were conducted using a wide range of alternative methods for two-way fixed-effects adjustments. In contrast with findings from previous studies, we found null effects on occupational fatal injuries, as well as on all other health outcomes. Overall, our results indicate that findings from previous studies are based on very thin empirical evidence, with potentially underestimated standard errors and unobserved confounders. Our results highlight the importance of revisiting research questions using updated methodological tools.

Leveraging Causal Inference Techniques for Robust Root Cause Identification in Complex Systems

Leveraging Causal Inference Techniques for Robust Root Cause Identification in Complex Systems

Causal effects of policy and occupant behavior on cooling energy

Rising energy consumption in buildings has prompted architectural engineers and policymakers to ask “what-if” questions in the pursuit of energy efficiency. Current simulation and machine learning methods rely primarily on correlations and associations necessitating causal inference techniques. In this research, a causal inferential approach combining double machine learning and domain knowledge using directed acyclic graphs is employed. The utility of this approach is demonstrated on the 2015 United States Residential energy consumption survey to evaluate the impact of energy policies and occupant behavior on cooling energy consumption. The results show that energy policies such as energy audits, proper insulation, access to interval data and Energy Star qualified windows significantly reduce energy use intensity (EUI). Additionally, air conditioning usage by occupants causally affects EUI. For instance, proper insulation reduces EUI by 5.603 kWh/m2 while changing static thermostat settings to automatic adjustments at certain times reduces EUI by 3.542 kWh/m2. This study suggests potential energy-saving policies for balanced building energy efficiency and functionality.

Game theory in the classroom: low cooperative relationships identify bullying patterns in elementary schools

Cooperation and bullying have a subtle yet important interaction that influences the social dynamics in elementary school classrooms. We investigate this interplay in a large sample of 1112 students across 47 public primary classrooms in Chile. Using a video game interface to create a dyadic, non-anonymous social dilemma, we map the cooperative social network within each classroom. In addition, we collect peer nomination data and use the Illinois Bullying Scale to categorize students as bullies, victims, or bully victims. Our results indicate that low levels of received cooperation significantly increase the likelihood of students being identified with the dual role of both bully and victim, known as the bully-victim profile. This negative relationship remains robust even after controlling for demographic and classroom context variables using multilevel regression models and is consistent when employing causal inference techniques such as statistical matching. We propose that the relationship between received cooperation and the bully-victim profile stems from the capacity of received cooperation to capture key factors influencing social relationships among students, such as popularity, prosociality, GPA, and aggressiveness. Our study contributes to the understanding of human interaction in educational settings and it offers a new framework for targeted interventions in primary education, providing insights for future educational policies and practices.

Artificial intelligence-empowered assessment of bile duct stone removal challenges

The aim of this investigation was to unravel the complexities inherent in endoscopic retrograde cholangiopancreatography (ERCP) procedures for bile duct stone removal by leveraging advanced artificial intelligence (AI) methodologies to support clinical decision-making and optimize patient outcomes. We introduced the Assessment of Stone Removal Challenges (ACRS) system, a novel integration of Data-efficient Image Transformers (DeiT) for image data analysis and eXtreme Gradient Boosting (XGBoost) for clinical data interpretation. Our study included a patient cohort of 2,129 individuals, focusing on training the ACRS system to achieve high diagnostic precision. Using logistic regression, we identified pivotal predictors affecting the complexity of bile duct stone removal. These findings were visually represented through forest plots and nomograms. Analytical and visualization processes were conducted using the Python and R programming languages, adhering to a p value significance threshold of less than 0.05. By utilizing DeiT enhanced by transfer learning and XGBoost for clinical data interpretation, the system achieved an accuracy of 0.83 and a perfect recall rate on a test set of 2,129 patients. Gradient-weighted class activation mapping (Grad-CAM) and SHapley Additive exPlanations (SHAP) provided in-depth diagnostic insights. Logistic regression was applied to identify crucial clinical predictors for stone removal difficulty, as visualized through forest plots and nomograms. These tools facilitated measurable assessments of procedural complexity, making significant strides in gastroenterology diagnostics and decision-making. By adopting causal inference techniques, the system effectively quantifies the influence of various clinical entities on stone removal difficulty, thereby augmenting both diagnostic precision and procedural strategies in ERCP.

The delayed and combinatorial response of online public opinion to the real world: An inquiry into news texts during the COVID-19 era

In sociological research based on online public opinion, scholars often overlook the delay and combinatory nature of online responses to real-world events. This study aims to explore the delayed and combinatory responses of online public opinion to the intensity of the COVID-19 pandemic. Specifically, we seek to answer the following questions: (a) Is there a temporal delay in the response of online public opinion to the intensity of the pandemic? (b) Does this delay exhibit general characteristics of social networks, such as combinatory effects and higher-order interactions? To address these questions, we employ natural language processing techniques to extract online public opinion data and utilize statistical and machine learning-based causal inference methods for analysis. The findings indicate that online public opinion’s response to the intensity of COVID-19 is not immediate but rather exhibits a long-term lag. Identical COVID-19 intensity data can trigger multiple delayed public opinion responses, while a single delayed public opinion datum may be influenced by multiple preceding COVID-19 intensity data points. This delayed response of online public opinion and its higher-order network characteristics result in a waveform structure of real-world impacts influenced by online public opinion. We also utilized machine learning causal inference techniques to investigate the sensitivity differences in online public opinion responses to COVID-19 during various time periods.

Unbiased news recommendation model combining time and content

The influence of popularity bias on news recommendation systems is substantial, as it increased the exposure of popular news and their being pushes to irrelevant users frequently, leading to less-than-ideal performance of the system. Existing popularity debiasing methods aim to optimize the fairness of recommendations so that unpopular news have more chances to be recommended. However, they ignore the timeliness that characterizes news, and some outdated news are recommended to users due to their low popularity. Unfortunately, users are not interested in outdated news. In addition, existing debiasing methods use content-independent popularity computation, which cannot accurately capture the dynamic changes in news popularity, thus reducing the effectiveness of debiasing. In this research, we introduce an innovative method named Time and Content-aware Causal Model, referred to as TCCM, which mitigates the popularity bias using current causal inference techniques that work well. The difference is that, first, we design a novel causal graph to analyze user interaction considering that news is timeliness. Based on this causal graph, TCCM models the influences of three factors on user interaction, namely, the timeliness of news, the popularity of news, and the matching between the content of news and the interest of the user, in order to solve the problem of recommending outdated news to users after de-biasing. Second, we consider the influences of news content on popularity and propose a new popularity estimation method. In TCCM, we improve the debiasing effect by combining news content (entity and word popularity) to obtain more accurate news popularity. Extensive experiments conducted on widely recognized benchmark datasets show that TCCM surpasses several cutting-edge approaches.

Causality-inspired crop pest recognition based on Decoupled Feature Learning.

Ensuring the efficient recognition and management of crop pests is crucial for maintaining the balance in global agricultural ecosystems and ecological harmony. Deep learning-based methods have shown promise in crop pest recognition. However, prevailing methods often fail to address a critical issue: biased pest training dataset distribution stemming from the tendency to collect images primarily in certain environmental contexts, such as paddy fields. This oversight hampers recognition accuracy when encountering pest images dissimilar to training samples, highlighting the need for a novel approach to overcome this limitation. We introduce the Decoupled Feature Learning (DFL) framework, leveraging causal inference techniques to handle training dataset bias. DFL manipulates the training data based on classification confidence to construct different training domains and employs center triplet loss for learning class-core features. The proposed DFL framework significantly boosts existing baseline models, attaining unprecedented recognition accuracies of 95.33%, 92.59%, and 74.86% on the Li, DFSPD, and IP102 datasets, respectively. Extensive testing on three pest datasets using standard baseline models demonstrates the superiority of DFL in pest recognition. The visualization results show that DFL encourages the baseline models to capture the class-core features. The proposed DFL marks a pivotal step in mitigating the issue of data distribution bias, enhancing the reliability of deep learning in agriculture. © 2024 Society of Chemical Industry.

The heterogeneous impact of population mobility on the influent characteristics of wastewater treatment facilities

Improving the resilience of wastewater treatment facilities (WWTFs) has never been more important with rising risks of disasters under climate change. Beyond physical damages, non-physical shocks induced by disasters warrant attention. Human mobility is a vital mediator in transferring the stresses from extreme events into tangible challenges for urban sewage systems by reshaping influent characteristics. However, the impact path remains inadequately explored. Leveraging the stay-at-home orders during the COVID-19 pandemic as a natural experiment, this study aims to quantify and interpret the heterogeneous impacts of mobility reduction on the influent characteristics of WWTFs with different socio-economic, infrastructural, and climatic conditions. To achieve this goal, we developed a research framework integrating causal inference and interpretable machine learning techniques. Based on the empirical data from China, we find that 79.1% of the studied WWTFs, typically located in cities with well-developed drainage infrastructures and low per capita water usage, exhibited resilience against drastic mobility reduction. In contrast, 20.9% of the studied WWTFs displayed significant variations in influent characteristics. Large-capacity WWTFs in subtropical regions encountered challenges with low-load operations, and small-capacity facilities in suburban areas grappled with nutrient imbalances. This study provides valuable insights to equip WWTFs in anticipating and adapting potential variations in influent characteristics triggered by mobility reduction.

Major Problems in Clinical Psychological Science and How to Address them. Introducing a Multimodal Dynamical Network Approach

Abstract Background Despite impressive dissemination programs of best-practice therapies, clinical psychology faces obstacles in developing more efficacious treatments for mental disorders. In contrast to other medical disciplines, psychotherapy has made only slow progress in improving treatment outcomes. Improvements in the classification of mental disorders could enhance the tailoring of treatments to improve effectiveness. We introduce a multimodal dynamical network approach, to address some of the challenges faced by clinical research. These challenges include the absence of a comprehensive meta-theory, comorbidity, substantial diagnostic heterogeneity, violations of ergodicity assumptions, and a limited understanding of causal processes. Methods Through the application of multimodal dynamical network analysis, we describe how to advance clinical research by addressing central problems in the field. By utilizing dynamic network analysis techniques (e.g., Group Iterative Multiple Model Estimation, multivariate Granger causality), multimodal measurements (i.e., psychological, psychopathological, and neurobiological data), intensive longitudinal data collection (e.g., Ecological Momentary Assessment), and causal inference methods (e.g., GIMME), our approach could improve the comprehension and treatment of mental disorders. Under the umbrella of the systems approach and utilizing e.g., graph theory and control theory, we aim to integrate data from longitudinal, multimodal measurements. Results The multimodal dynamical network approach enables a comprehensive understanding of mental disorders as dynamic networks of interconnected symptoms. It dismantles artificial diagnostic boundaries, facilitating a transdiagnostic view of psychopathology. The integration of longitudinal data and causal inference techniques enhances our ability to identify influential nodes, prioritize interventions, and predict the impact of therapeutic strategies. Conclusion The proposed approach could improve psychological treatment by providing individualized models of psychopathology and by suggesting individual treatment angles.

Causal Latent Class Analysis with Distal Outcomes: A Modified Three-Step Method Using Inverse Propensity Weighting

Bias-adjusted three-step latent class (LC) analysis is a popular technique for estimating the relationship between LC membership and distal outcomes. Since it is impossible to randomize LC membership, causal inference techniques are needed to estimate causal effects leveraging observational data. This paper proposes two novel strategies that make use of propensity scores to estimate the causal effect of LC membership on a distal outcome variable. Both strategies modify the bias-adjusted three-step approach by using propensity scores in the last step to control for confounding. The first strategy utilizes inverse propensity weighting (IPW), whereas the second strategy includes the propensity scores as control variables. Classification errors are accounted for using the BCH or ML corrections. We evaluate the performance of these methods in a simulation study by comparing it with three existing approaches that also use propensity scores in a stepwise LC analysis. Both of our newly proposed methods return essentially unbiased parameter estimates outperforming previously proposed methods. However, for smaller sample sizes our IPW based approach shows large variability in the estimates and can be prone to non-convergence. Furthermore, the use of these newly proposed methods is illustrated using data from the LISS panel.

The effect of fire-induced forest-degradation on rainfall: A causal inference analysis of the case of the Brazilian Amazon

Forest degradation by fires is growing in a fast pace in the Brazilian portion of the Amazonian rainforest, damaging the regional environment and economy. Seeking to measure the impact of such process on rainfall, the paper analyses a wide diversity of fine-grained satellite measurements with causal inference techniques. Results show that agricultural fires increase forest fires in 40 % of the usual rise in forest fires due to seasonal warming. Rainfall is reduced in 25 % of the rainfall decay normally caused by seasonal drying. Impacts on agricultural production were detected, but the evidence was not strong. These findings considerably expand the hitherto available evidence, which is not only scarce and biased by confounders, but essentially uni-disciplinary and thus partial in terms of the acknowledged consequences.

Prognostic impact and causality of age on oncological outcomes in women with endometrial cancer: a multimethod analysis of the randomised PORTEC-1, PORTEC-2, and PORTEC-3 trials

Numerous studies have shown that older women with endometrial cancer have a higher risk of recurrence and cancer-related death. However, it remains unclear whether older age is a causal prognostic factor, or whether other risk factors become increasingly common with age. We aimed to address this question with a unique multimethod study design using state-of-the-art statistical and causal inference techniques on datasets of three large, randomised trials. In this multimethod analysis, data from 1801 women participating in the randomised PORTEC-1, PORTEC-2, and PORTEC-3 trials were used for statistical analyses and causal inference. The cohort included 714 patients with intermediate-risk endometrial cancer, 427 patients with high-intermediate risk endometrial cancer, and 660 patients with high-risk endometrial cancer. Associations of age with clinicopathological and molecular features were analysed using non-parametric tests. Multivariable competing risk analyses were performed to determine the independent prognostic value of age. To analyse age as a causal prognostic variable, a deep learning causal inference model called AutoCI was used. Median follow-up as estimated using the reversed Kaplan-Meier method was 12·3 years (95% CI 11·9-12·6) for PORTEC-1, 10·5 years (10·2-10·7) for PORTEC-2, and 6·1 years (5·9-6·3) for PORTEC-3. Both overall recurrence and endometrial cancer-specific death significantly increased with age. Moreover, older women had a higher frequency of deep myometrial invasion, serous tumour histology, and p53-abnormal tumours. Age was an independent risk factor for both overall recurrence (hazard ratio [HR] 1·02 per year, 95% CI 1·01-1·04; p=0·0012) and endometrial cancer-specific death (HR 1·03 per year, 1·01-1·05; p=0·0012) and was identified as a significant causal variable. This study showed that advanced age was associated with more aggressive tumour features in women with endometrial cancer, and was independently and causally related to worse oncological outcomes. Therefore, our findings suggest that older women with endometrial cancer should not be excluded from diagnostic assessments, molecular testing, and adjuvant therapy based on their age alone. None.

Intelligent Agents and Causal Inference: Enhancing Decision-Making through Causal Reasoning

This study examines the incorporation of causal inference methods into intelligent entities and examines the benefits of utilizing causal reasoning to improve decision-making procedures. This study entails conducting an experimental evaluation within a video game setting to evaluate the performance of three separate agent types: ExplorerBOT, GuardBOT, and CausalBOT. The ExplorerBOT utilizes a stochastic path selection technique for task completion, whereas the GuardBOT remains immobile yet exhibits exceptional proficiency in identifying and neutralizing other bots. On the other hand, the CausalBOT utilizes sophisticated causal inference methods to examine the underlying factors contributing to the failures noticed in the task completion of the ExplorerBOT. The aforementioned feature allows CausalBOT to make informed decisions by selecting paths that have a greater likelihood of achieving success. The main purpose of these experiments is to assess and compare the effectiveness of two distinct bots, namely ExplorerBOT and CausalBOT, in accomplishing their respective objectives. To facilitate comparison, two iterations of the ExplorerBOT are utilized. The initial iteration is predicated exclusively on stochastic path selection and necessitates a more profound understanding of the variables that impact the achievement of tasks. On the other hand, the second version integrates an algorithm for informed search. In contrast, CausalBOT employs causal inference techniques to discover the underlying causes of failures exhibited by ExplorerBOTs and collect pertinent data. Through the process of discerning the fundamental causal mechanisms, CausalBOT is able to make well-informed decisions by selecting pathways that maximize the probability of successfully completing a given job. The utilization of this approach greatly boosts the decision-making powers of CausalBOT, hence enabling it to effectively adapt and overcome problems in a more efficient manner when compared to alternative agents.

Reinforcement learning-based SDN routing scheme empowered by causality detection and GNN.

In recent years, with the rapid development of network applications and the increasing demand for high-quality network service, quality-of-service (QoS) routing has emerged as a critical network technology. The application of machine learning techniques, particularly reinforcement learning and graph neural network, has garnered significant attention in addressing this problem. However, existing reinforcement learning methods lack research on the causal impact of agent actions on the interactive environment, and graph neural network fail to effectively represent link features, which are pivotal for routing optimization. Therefore, this study quantifies the causal influence between the intelligent agent and the interactive environment based on causal inference techniques, aiming to guide the intelligent agent in improving the efficiency of exploring the action space. Simultaneously, graph neural network is employed to embed node and link features, and a reward function is designed that comprehensively considers network performance metrics and causality relevance. A centralized reinforcement learning method is proposed to effectively achieve QoS-aware routing in Software-Defined Networking (SDN). Finally, experiments are conducted in a network simulation environment, and metrics such as packet loss, delay, and throughput all outperform the baseline.

A Personalized Collaborative Filtering Recommendation System Based on Bi-Graph Embedding and Causal Reasoning.

The integration of graph embedding technology and collaborative filtering algorithms has shown promise in enhancing the performance of recommendation systems. However, existing integrated recommendation algorithms often suffer from feature bias and lack effectiveness in personalized user recommendation. For instance, users' historical interactions with a certain class of items may inaccurately lead to recommendations of all items within that class, resulting in feature bias. Moreover, accommodating changes in user interests over time poses a significant challenge. This study introduces a novel recommendation model, RCKFM, which addresses these shortcomings by leveraging the CoFM model, TransR graph embedding model, backdoor tuning of causal inference, KL divergence, and the factorization machine model. RCKFM focuses on improving graph embedding technology, adjusting feature bias in embedding models, and achieving personalized recommendations. Specifically, it employs the TransR graph embedding model to handle various relationship types effectively, mitigates feature bias using causal inference techniques, and predicts changes in user interests through KL divergence, thereby enhancing the accuracy of personalized recommendations. Experimental evaluations conducted on publicly available datasets, including "MovieLens-1M" and "Douban dataset" from Kaggle, demonstrate the superior performance of the RCKFM model. The results indicate a significant improvement of between 3.17% and 6.81% in key indicators such as precision, recall, normalized discount cumulative gain, and hit rate in the top-10 recommendation tasks. These findings underscore the efficacy and potential impact of the proposed RCKFM model in advancing recommendation systems.