User Interaction Data Research Articles

An AI-Enabled Framework for Mood Analysis and Mental Health Support Using User Interaction Data

The prevalence of mental health challenges such as anxiety, depression, and stress is steadily increasing, yet traditional diagnostic methods often remain limited in terms of accessibility, personalization, and the ability to provide real-time insights. To address these gaps, an AI-driven mental health diagnostic platform, has been developed to offer a more effective, data-centric solution. By leveraging advanced machine learning algorithms and behavioral analytics, this platform assesses users’ mental well-being with precision and responsiveness. The platform integrates mood pattern analysis, user interaction monitoring, and self-reported data to deliver personalized diagnostics and tailored mental health recommendations. Through actionable insights, coping strategies, and curated mental health resources, it empowers users to better understand and manage their mental health proactively. Designed for accessibility and ease of use, this framework represents a significant step toward redefining mental health care with a user-centric, real-time, and technology-enabled approach. Key Words: Mental Health Diagnostics, Artificial Intelligence, Behavioral Analytics, Mood Pattern Analysis, Personalized Recommendations,

Research on Advertising Effect Evaluation Model and Algorithm Based on Multimodal Data

The rapid growth of digital advertising has created a dynamic situation with a wide range of content types and complex customer behavior. The variety of data sources accessible in the age of digital marketing has made assessing the success of advertising efforts a difficult task. This study’s goal is to explore a novel multimodal data-based advertising effect evaluation model and algorithm to offer a thorough examination of digital advertising effectiveness. To collect multi-model data, which include text, audio, video, and user interaction data related to the advertisement, the data were preprocessed to handle stemming, image resizing, and normalization of the obtained multi-model data. Then, the features are extracted using convolution neural networks (CNNs) employed to extract visual features, while bag of words (BoW) is used to analyze audio and text data for emotional patterns. Finally, the extracted features are fused. This study proposed a novel modified crayfish-optimized multimodal long short-term memory (MCFO-MLSTM) to evaluate the effectiveness of the advertisement. The model dynamically adjusts video content and playback mode based on user emotions and behaviors, enhancing engagement and ad effectiveness. The findings show that the suggested approach performs better in terms of consumer satisfaction with tailored advertisements, enhances advertising design, and increases viewer engagement. This research contributes to advancing advertising evaluation techniques by incorporating multimodal data and emotion-driven algorithms.

Multi-behavior aware recommendation with joint contrastive learning and reinforced negative sampling

Traditional recommendation systems usually rely on single user-item interaction information to capture the characteristics of users and items. However, in real-world applications, the interactions between users and items are far more diverse. Therefore, efficiently integrating multidimensional interaction data is crucial to improve the performance of recommendation systems. The existing methods still face challenges in terms of modeling behavioral dependencies, handling skewed data distributions, and optimizing the quality of negative samples. To address these obstacles, a joint contrastive learning and reinforced negative sampling model is proposed to multi-behavior aware recommendation. First, user interaction data concerning multiple behaviors are incorporated to automatically capture and encode behavioral context information and high-order dependencies, enabling a more comprehensive understanding of user behavior patterns. Second, contrastive learning is applied to the user data acquired under different behaviors to construct enhanced personalized user behavior representations. Additionally, a similarity-based strategy using the dot product operation is introduced to identify hard negative samples with high similarity to users, thereby strengthening the item-side representations. The model is optimized through multi-task learning. Extensive experiments conducted on three real datasets, i.e., IJCAI, Tmall, and Retail, demonstrate that the proposed method improves upon the performance of the existing multi-behavior baseline models by 14.5%, 6.0%, and 1.1% in terms of hit ratio and by 19.7%, 9.4%, and 2.3% in terms of the normalized discounted cumulative gain, respectively, verifying the effectiveness of the new approach.

Optimizing AIGC-Driven Visual Communication: A Deep Learning Framework for Adaptive Image Generation and User-Centered Content Scheduling

The rapid development of AIGC technology has opened a new avenue for personalized visual communication. This study builds a deep learning framework that integrates adaptive image generation and user-driven content arrangement, aiming to improve the expressiveness and delivery efficiency of visual creation. The technical architecture consists of two modules: the image generation engine based on Stable Delivery and the content scheduling system driven by the GRU algorithm. The two modules are connected by a feedback loop and can dynamically optimize image quality and push timing based on user interaction data. Experiments based on the LAION-400M dataset and real user logs show that this system performs remarkably well in indicators such as image fidelity and semantic fit, and the click-through rate and user satisfaction score have significantly improved. Compared with the static scheduling system, the system achieved an FID score of 12.4, and the average click-through rate increased by 18.7%. This solution not only promotes technological innovation in AIGC, but also provides practical tools for scenarios such as brand communication and digital marketing in the new media environment.

TeleScope A Longitudinal Dataset for Investigating Online Discourse and Information Interaction on Telegram

Telegram is a globally popular instant messaging platform known for its strong emphasis on security, privacy, and unique social networking features. It has recently emerged as the host for various cross-domain analysis and research works, such as social media influence, propaganda studies, and extremism. This paper introduces TeleScope, an extensive dataset suite that, to our knowledge, is the largest of its kind. It comprises metadata for about 500K Telegram channels and downloaded message metadata for about 71K public channels, accounting for around 120M crawled messages. We also release channel connections and user interaction data built using Telegram’s message-forwarding feature to study multiple use cases, such as information spread and message-forwarding patterns. In addition, we provide data enrichments, such as language detection, active message posting periods for each channel, and Telegram entities extracted from messages, that enable online discourse analysis beyond what is possible with the original data alone. The dataset is designed for diverse applications, independent of specific research objectives, and sufficiently versatile to facilitate the replication of social media studies comparable to those conducted on platforms like X (formerly Twitter).

Recommending art online: investigating user engagement and interactions with a digital collection

ABSTRACT Museum online collections now contain millions of objects, making developing tools for supporting users in navigating these a priority. We present a user-centric study of a recommendation system created to browse Art UK’s digital collection according to personal preference. Three forms of recommendations were explored: image-only, metadata-only, and a combination of the two. Recommendations were evaluated against a baseline of artworks chosen at random. To create a holistic picture of user engagement, both subjective ratings and interaction data were collected. Whilst ratings data showed a uniformly positive perception of the system, the interaction data showed significantly deeper and longer engagement with the artworks that were recommended. Results indicate (1) that recommendations can be a useful tool to help users explore online art collections, and (2) that when evaluating recommendation systems, interaction data can capture patterns of increased engagement that are not evident in subjective ratings.

An AI and Machine Learning Approach to Precision Targeting in Social Media Advertising

As social media has become a leading platform for digital marketing, businesses are increasingly focusing on personalized advertising strategies. One of the most effective approaches involves applying artificial intelligence (AI) and machine learning (ML) to optimize ad targeting across social media platforms. This approach typically involves several key stages: data collection, data preprocessing, model training, and optimization of targeting strategies. In the data collection phase, user interaction and behavioral data are gathered from various online sources. This data is then preprocessed by removing noise, filling in missing values, and converting categorical features into numerical formats suitable for machine learning models. Once the data is cleaned, various ML algorithms such as logistic regression, decision trees, and neural networks are trained to identify patterns between user features and ad engagement outcomes. After training, model performance is evaluated to determine the most effective algorithm. The best-performing model is then deployed to predict outcomes such as the likelihood of a user clicking on an ad. These predictions can be used to continuously refine targeting strategies based on new data, making the system adaptive and increasingly accurate over time. Using ML to optimize social media ad targeting offers several advantages. It helps businesses achieve better return on investment (ROI) through more precise targeting, reduces manual workload, and enables scalable, automated ad delivery. This framework provides a data-driven method for enhancing the effectiveness of digital advertising campaigns.

DRLAttack: A Deep Reinforcement Learning-Based Framework for Data Poisoning Attack on Collaborative Filtering Algorithms

Collaborative filtering, as a widely used recommendation method, is widely applied but susceptible to data poisoning attacks, where malicious actors inject synthetic user interaction data to manipulate recommendation results and secure illicit benefits. Traditional poisoning attack methods require in-depth understanding of the recommendation system. However, they fail to address its dynamic nature and algorithmic complexity, thereby hindering effective breaches of the system’s defensive mechanisms. In this paper, we propose DRLAttack, a deep reinforcement learning-based framework for data poisoning attacks. DRLAttack can launch both white-box and black-box data poisoning attacks. In the white-box setting, DRLAttack dynamically tailors attack strategies to recommendation context changes, generating more potent and stealthy fake user interactions for the precise targeting of data poisoning. Furthermore, we extend DRLAttack to black-box settings. By introducing spy users to simulate the behavior of active and inactive users into the training dataset, we indirectly obtain the promotion status of target items and adjust the attack strategy in response. Experimental results on real-world recommendation system datasets demonstrate that DRLAttack can effectively manipulate recommendation results.

Learning Human Feedback from Large Language Models for Content Quality-aware Recommendation

Recommender systems are widely employed to mitigate information overload by tailoring online content to individual preferences. Existing recommendation methods typically focus on optimizing the relevance between candidate item content and user historical behaviors. However, these methods often neglect the quality of recommended content, which can negatively affect user experience and hinder the long-term growth of platforms. In fact, addressing this issue is particularly challenging, as signal on content quality feedback is typically sparse in the user interaction data (e.g., clicks) commonly used for model training. In this article, we propose a human feedback alignment framework for recommender system (HFAR), which leverages well-aligned large language models to simulate human feedback on content quality to enhance recommendation. Specifically, we propose a multi-task learning-based knowledge transfer framework to infuse recommendation models with an awareness of fine-grained feedback on content quality from targeted perspectives. Furthermore, we develop a contrastive learning-based feedback integration mechanism to embed targeted human feedback into the ranking strategy to enable quality-aware recommendation decision-making. Besides, we propose a multi-objective joint training framework to optimize the model jointly under utility and quality objectives. Experiments show that HFAR achieves a maximum improvement of 84.78% in recommendation quality, while maintaining both recommendation accuracy and efficiency.

Quantification and Analysis of Group Sentiment in Electromagnetic Radiation Public Opinion Events

This research focuses on developing a sentiment-based system to analyze public opinion on electromagnetic radiation in online networks. Issues related to EMR, such as the NIMBY effect and negative public sentiment, can lead to health crises, social conflicts, and challenges in decision-making. This study addresses limitations in existing research, including inaccurate data collection and a lack of systematic analysis. By incorporating Jieba Chinese word segmentation technology, this study introduces an innovative data collection method based on topic similarity, significantly improving data accuracy. Additionally, this research establishes a comprehensive public opinion analysis framework that integrates user follower counts, geographical distribution, and interaction data. This framework facilitates the identification of sources of negative sentiment and the development of effective response strategies. As a case study, the dissemination patterns of EMR-related public opinion on Weibo are analyzed, focusing on group sentiment and social interaction. The proposed system achieves a 65.85% improvement in data collection accuracy, demonstrating its effectiveness. Furthermore, this study provides actionable recommendations for relevant departments and governments to monitor, analyze, and respond to EMR-related public opinion. By enhancing decision-making and protecting public interests, this study highlights the role of technology in improving social governance and substantial development.

From Browsing to Buying: The Role of Augmented Reality in Shortening The Customer Journey in E-Commerce

Objective: This research aims to explore how Augmented Reality (AR) technology contributes to shortening the customer journey in e-commerce by enhancing product exploration, facilitating evaluation, and accelerating purchase decisions. The focus is placed on industries where product visualization is critical, namely cosmetics and furniture. Research Design & Methods: A qualitative approach was used, involving a comprehensive literature analysis and in-depth interviews with ten active users of an AR-integrated e-commerce platform. These users were purposively selected from two sectors-cosmetics and furniture-to gain insight into real user experiences across different product types. Thematic analysis was used to identify recurring patterns and behavioral tendencies. Findings: The study found that AR significantly impacts three key stages of the customer journey: product exploration, evaluation of alternatives, and final decision-making. AR increases user confidence through realistic visualizations, reduces perceived risk by providing interactive simulations, and increases engagement through control and enjoyment. In addition, AR contributes to the value of the experience, leading to greater confidence and purchase intention. Implications & Recommendations: From a managerial perspective, integrating AR into e-commerce platforms creates strategic value by shortening the buying cycle, enriching the customer experience, and strengthening brand loyalty. Businesses are advised to invest in user-friendly AR interfaces and personalize AR experiences based on user interaction data. Future research should expand to other industries and incorporate quantitative methods to validate and generalize the findings. Contribution & Value Added: This study contributes to the literature by positioning AR not only as a visualization tool but also as a strategic enabler to accelerate consumer decision-making. It adds value by bridging the gap between immersive technologies and practical marketing applications in the digital economy.

Healthify: A Conversational AI for Mental Health Support Using Groq and LangChain Frameworks

Mental health issues are a growing global concern, with many individuals facing barriers to accessing timely and effective care. Recent advancements in artificial intelligence (AI) have opened new possibilities for providing continuous, personalized emotional support. This study presents Healthify, an AI-powered mental health support chatbot designed to provide empathetic, context-aware conversations for individuals seeking emotional guidance. By leveraging LangChain and the Groq API, Healthify combines state-of-the-art natural language processing (NLP) models with a robust memory system, ensuring consistent and personalized interactions tailored to the user’s needs. The chatbot’s ability to engage users in continuous dialogues aims to bridge gaps in mental health care, offering an accessible and non-judgmental platform for support. This research evaluates the effectiveness of Healthify in providing emotional support by analysing its ability to simulate empathetic conversation and address various mental health concerns. Through user interaction data and qualitative analysis, we assess the chatbot’s responsiveness, emotional intelligence, and potential impact on mental well- being. The findings demonstrate that AI-driven platforms can provide a valuable supplement to traditional mental health resources, particularly for individuals who may be reluctant or unable to seek professional help. We conclude that Healthify offers a promising direction for integrating AI in mental health interventions, with implications for expanding access to mental health support on a global scale.

Application of Reinforcement Learning Techniques in Software Testing of Android Applications: A Systematic Literature Review

Software testing is a critical process in ensuring the quality and reliability of applications before they are deployed to production. However, it is resource-intensive and often tedious, particularly in the context of Android applications, which pose unique challenges due to their vast state space, diverse user interactions, and variable behaviors. Reinforcement learning (RL), a machine learning framework where agents interact with environments to improve decision-making policies, has gained attention for its potential in software testing. This systematic literature review examines the application of reinforcement learning in software testing of Android applications, focusing on widely researched areas, prevalent techniques, and emerging trends. The review analyzes 22 selected studies from an initial pool of over 30,000 articles published between 2020 and 2024. The findings highlight that automated testing is the primary focus in this domain, with Q-learning emerging as the dominant RL technique. Actor-critic methods, deep Q-networks (DQN), and policy gradient approaches are also explored in several studies, aiming to improve the adaptability and efficiency of testing processes. Most research emphasizes fault detection and coverage maximization, often targeting event-driven interactions and GUI-based behaviors. Despite significant advancements, the study identifies underexplored areas, such as test case prioritization and the integration of user behavior or user interaction data, as promising directions for future research. This review contributes to understanding the current landscape and offers guidance for future RL-based Android application testing investigations.

Machine learning approach for optimizing usability of healthcare websites

In today’s digital era, hospital websites serve as crucial informational resources, providing patients with easy access to medical services. Ensuring the usability of these websites is essential, as it directly impacts users’ ability to navigate and retrieve vital medical information. Despite the recognized importance of website usability in the healthcare sector, there is a notable lack of empirical studies leveraging machine learning to assess this usability. This study aims to fill this gap by evaluating the user-friendliness of hospital websites using machine learning models, including Decision Trees, Random Forests, Ridge Regression, and Support Vector Regression. The dataset used in this analysis was generated by a custom-built automated tool developed by the author, which assessed the usability of 100 hospital websites. Among the models, Random Forest Regression and Ridge Regression demonstrated exceptional performance, achieving an accuracy rate of 98% and 87% respectively. Key metrics such as R-square, Mean Square Error (MSE), Mean Absolute Error (MAE), and Explained Variance Score (EVS) confirmed the model’s predictive accuracy. Cross-validation further highlighted the robustness of Ridge Regression exhibiting low overfitting. The importance rankings consistently underscored the critical role of overall usability in predicting website performance. The study recommends expanding the dataset to include a broader range of healthcare websites, integrating user interaction data, and exploring advanced analytical techniques like deep learning to enhance future usability assessments. These advancements could lead to optimized website design and improved digital healthcare experiences for a diverse range of users.

Decoding Global Palates: Unveiling Cross-Cultural Flavor Preferences Through Online Recipes.

Navigating cross-cultural food choices is complex, influenced by cultural nuances and various factors, with flavor playing a crucial role. Understanding cultural flavor preferences helps individuals make informed food choices in cross-cultural contexts. We examined flavor differences across China, the US, and Germany, as well as consistent flavor preference patterns using online recipes from prominent recipe portals. Distinct from applying traditional food pairing theory, we directly mapped ingredients to their individual flavor compounds using an authorized database. This allowed us to analyze cultural flavor preferences at the molecular level and conduct machine learning experiments on 25,000 recipes from each culture to reveal flavor-based distinctions. The classifier, trained on these flavor compounds, achieved 77% accuracy in discriminating recipes by country in a three-class classification task, where random choice would yield 33.3% accuracy. Additionally, using user interaction data on appreciation metrics from each recipe portal (e.g., recipe ratings), we selected the top 10% and bottom 10% of recipes as proxies for appreciated and less appreciated recipes, respectively. Models trained within each portal discriminated between the two groups, reaching a maximum accuracy of 66%, while random selection would result in a baseline accuracy of 50%. We also explored cross-cultural preferences by applying classifiers trained on one culture to recipes from other cultures. While the cross-cultural performance was modest (specifically, a max accuracy of 54% was obtained when predicting food preferences ofthe USusers with models trained on the Chinesedata), the results indicate potential shared flavor patterns, especially between Chinese and US recipes, which show similarities, while German preferences differ. Exploratory analyses further validated these findings: we constructed ingredient networks based on co-occurrence relationships to label recipes as savory or sweet, and clustered the flavor profiles of compounds as sweet or non-sweet. These analyses showed opposing trends in sweet vs. non-sweet/savory appreciation between US and German users, supporting the machine learning results. Although our findings are likely to be influenced by biases in online data sources and the limitations of data-driven methods, they may still highlight meaningful cultural differences and shared flavor preferences. These insights offer potential for developing food recommender systems that cater to cross-cultural contexts.

DOGE: LLMs-Enhanced Hyper-Knowledge Graph Recommender for Multimodal Recommendation

In recent years, there has been a burgeoning interest in multimodal recommender systems within the recommendation systems domain. These systems aim to understand user preferences by leveraging both user interaction data and multimodal information associated with items. This approach frequently results in superior recommendation accuracy compared to traditional models that rely solely on user-item interactions. Despite the advancements of these methods, there is a relatively low utilization of image features in propagating item-item characteristics, an overreliance on text feature similarity, and a frequent neglect of the deep relationships between items, users, and modalities. In response to these challenges, we introduce a novel model termed LLMs-Enhanced Hyper-Knowledge Graph Recommender for Multimodal Recommendation (DOGE). DOGE utilizes large language models (LLMs) to understand image information under the guidance of text information, generating cross-modal features that effectively enhance the relationship between text and image modalities. Subsequently, DOGE constructs a Hyper-Knowledge Graph (HKG) using user-item interaction information and modality features enhanced by large language models. This graph encompasses a wide range of item-item and user-user binary relations and hyper-relations, effectively expanding the feature propagation mechanisms and mitigating the overreliance on text modality. By learning on heterogeneous user-item graphs and homogeneous item-item, user-user graphs, DOGE enhances potential effective propagation between item features and user features, acquiring more effective feature representations of users and items. Comprehensive experimentation across three public real-world datasets illustrates that DOGE attains state-of-the-art (SOTA) performance, exhibiting a 7.2% improvement over the strongest baseline.

Assessing the Supply–Demand Matching and Spatial Flow of Urban Cultural Ecosystem Services: Based on Geospatial Data and User Interaction Data

Cultural ecosystem services (CESs) reflect the interaction between ecosystems and human well-being. Owing to constraints in data availability and existing methodological limitations, deriving information from non-material ecosystem attributes was inadequate. We took Yulin City, located in the northern Shaanxi Loess Plateau, as a case study. Based on open-source geospatial data and user interaction data from social media, a coupled multi-source model was applied to elucidate the spatial distribution of CESs’ supply–demand flow. The Maxent and LDA model were utilized to quantify CES supply–demand, whereas the breakpoint and gravity model were applied to explain the direction and intensity of CES flow. The results indicated the following: (1) aesthetic was the most perceivable CES in Yulin, with 27% high supply areas and four demand topics. And the perception of the educational CES was the least pronounced, with only 2% of high supply areas and two demand topics. (2) Yulin exhibited a notable mismatching in CES supply–demand, with the supply–demand matching area constituting only approximately 10%. In the center of the city, CESs displayed a spatial pattern of a supply–demand deficit, while areas farther from the city center presented a spatial pattern of a supply–demand surplus. (3) The flow of CESs followed a pattern of movement from peripheral counties to central counties and from less developed counties to more developed counties. We proposed the following targeted recommendations: introducing low-perception CESs to promote the enhancement of ecosystem services (ESs); and alleviating CES supply–demand mismatches by enhancing transportation accessibility and protecting the ecological environment. Simultaneously, attention should be directed towards the developmental disparities between counties, providing differentiated guidance for CES spatial flow. Our study provided a theoretical foundation for understanding CES supply–demand flow and offered scientific insights for the spatial development of urban CES.

Collaborative singular value decomposition with user-item interaction expansion for first-time user and item recommendations

In today's digital landscape, recommendation systems are essential for delivering personalized content and improving user engagement across various platforms. However, a key challenge known as the cold-start problem—where limited user-item interaction data hampers the ability to generate accurate recommendations—remains a significant obstacle, particularly for new users and items. To address this issue, this paper introduces an enhanced methodology combining collaborative singular value decomposition (Co-SVD) with an innovative approach to reduce data sparsity. The objective of this research is to improve recommendation accuracy in sparse data environments by leveraging collaborative information in the user-item interaction matrix. Extensive experiments conducted on an e-commerce dataset validate the superiority of the proposed Enhanced Co-SVD model over traditional Co-SVD, content-based filtering, and random recommendation methods across multiple metrics. Our approach demonstrates particular strength in cold-start scenarios, providing precise recommendations with minimal user interaction data. These findings have important implications for e-marketing, personalized user experiences, and overall business success in online environments.

MERN Stack-Based Job Portal: Design and Development

Full-stack development is crucial for building dynamic and scalable web applications, ensuring seamless user interaction and efficient data management. This project focuses on developing a full-stack job portal using the MERN (MongoDB, Express.js, React, Node.js) stack. The application facilitates job seekers in searching for opportunities and apply- ing for jobs, while employers can post job listings and manage applications efficiently. The backend is developed using Express.js and MongoDB, ensuring secure API endpoints and seamless data handling. The frontend is built with React.js, providing an interactive and user-friendly interface for job listings, applications, and user authentication. The platform integrates JWT-based authenti- cation for secure access, role-based access control (RBAC) for different user types (job seekers and employers), and real-time job application tracking. Key features include job search with filtering options, an interactive dashboard for applicants and recruiters, and profile management. Future enhancements aim to improve system scalability, incorporate AI-based job recommendations, and integrate third-party authentication and payment systems for premium job postings. Index Terms: Full-Stack Development, MERN Stack, Job Portal, React.js, Node.js, MongoDB, Express.js, JWT Authentication, Role-Based Access Control (RBAC)

Automatic Classification and Recommendation for English Teaching Materials Based on Natural Language Processing

The wide range of digital educational resources calls for developing an accurate and efficient method for categorizing and recommending English teaching materials. An automatic classification and recommendation system has been created and implemented using Natural Language Processing (NLP) techniques. Data from essays produced by English Language Learners (ELLs) in grades 8 through 12, as well as components including content, competency levels and score notes, are organized for this study. Models for precise language proficiency assessments are planned to be developed to enhance automated feedback mechanisms. Preprocessing methods such as stop-word removal, stemming and tokenization were applied to tidy up the data. Term Frequency–Inverse Document Frequency (TF–IDF) and word embeddings were two strategies used in the feature extraction process to convert textual data into numerical vectors. Then, the recently created Support Vector Machine–Neural Network–Genetic Algorithm (SVM–NN–GA) was fed to classify these vectors. The model’s performance was evaluated using F1-measure, accuracy, precision and recall metrics. Methods of collaborative filtering and content-based filtering were studied for the recommendation system. In contrast to collaborative filtering, which used user interaction data to identify patterns and suggest relevant items, content-based filtering matched materials with user preferences based on attributes gathered from NLP models. A hybrid recommendation system combines different approaches, increasing recommendations’ personalization and relevance. The results demonstrate that the hybrid recommendation and NLP-based categorization approach as a combination method suggestively improves the effectiveness of selecting appropriate teaching materials, helping teachers to enhance the learning process.