Retraction Note: Toward an intelligent tourism recommendation system based on artificial intelligence and IoT using Apriori algorithm

Retraction Note: A personalized recommendation system for teaching resources in sports using fuzzy C-means clustering technique

Personalized recommendation systems have become increasingly important in supporting intelligent decisionmaking in digital platforms, particularly in domains involving frequent and repetitive purchases such as household items. Unlike traditional recommendation scenarios that rely on explicit user ratings, household consumption is largely driven by implicit behavior patterns and usage frequency, making preference prediction more challenging. This paper presents an intelligent household item recommendation system based on the DSUM algorithm, which dynamically analyzes user–item interaction data to identify similarity patterns and generate personalized recommendations. The proposed approach utilizes structured transactional data to model user behavior and adapt recommendations according to evolving consumption needs. Performance evaluation is carried out using standard recommendation metrics, including accuracy, precision, recall, and F1-score, to assess the effectiveness of the system. Experimental results indicate that the DSUM-based model produces relevant and consistent recommendations while maintaining interpretability and computational efficiency, making it suitable for small to medium-scale household recommendation applications.

The application of artificial intelligence algorithms in news media systems is gradually forming an intelligent content ecosystem driven by deep models. We have constructed a multi module intelligent system that integrates Transformer, BERT, and knowledge graph, covering three major functions: semantic parsing, content recommendation, and intelligent generation. Based on multimodal embedding and Prompt optimization techniques, we have completed algorithm iterations on the basis of traditional CNN and LSTM architectures. The experiment showed that the BART model improved the BLEU score to 36.42 in the abstract generation task, which was 16.4% higher than that of Transformer; The Transformer in the recommendation system NDCG@10 Reaching 0.873, better than XGBoost and LSTM. The system exhibits significant advantages in both accuracy and response speed.

Artificial intelligence (AI) is also becoming an acknowledged game-changer in the world of agriculture, and it has the potential to increase productivity, efficiency, and food security. This paper discussed how AI can transform agriculture in Nigeria, its present use, opportunities, challenges and the possible effects. Qualitative research design was used, and semi-structured interviews with major stakeholders of active Nigerian agribusiness firms using AI were used. The companies offered their experience with AI-driven tools such as chatbots, predictive analytics, recommender systems, and pest detection models. The results showed that the early adoption of AI is transforming the agricultural practices by enhancing crop monitoring, improving farmer-market connections, optimising the use of inputs, and minimizing losses after harvesting. The companies cited positive effects of AI on productivity, farmer earnings, and rural development, and it was considered essential in enhancing food security and financial inclusion. But still, there were recurring issues that were found such as inadequate digital infrastructure, disjointed datasets, lack of digital literacy, and high implementation expenses. To solve these, companies implemented strategies like incremental implementation, open source technologies, offline solutions and collaborations with research institutions. The report concludes that AI can revolutionize the agricultural industry in Nigeria and lead to a tremendous economic growth, as long as systemic obstacles are overcome. Strategies to improve rural digital infrastructure, building centralised agricultural data systems, improving digital literacy, establishing inclusive financing systems, and collaboration between government, academia and the business sector are all recommended. These results can play a role in the continued discussion of digital agriculture and can serve as a means to implement AI usage in Sub-Saharan Africa on a larger scale.

For making well‑founded decisions on the elimination of failures and malfunctions occurring in railway infrastructure facilities, prompt access to information on previously identified faults and the dynamics of the irresolution is essential. Inspection logs such as DU‑46 contain valuable data on the condition of these facilities (tracks, turnouts, signals, power supply, contact lines, etc.); however, they are hardly used in practice when analyzing the causes of newly emerging failures. Aim. To develop an algorithm for processing DU‑46 log records that allows operators, up on request, to obtain information on previous malfunctions or maintenance activities on specific infrastructure objects. Methods. Text preprocessing, lemmatization using M. Korobov’s morphological analyzer, frequency‑based text analysis, TF–IDF, L2 normalization, cosine similarity calculation, and result sorting. Result. A prototype application has been developed that enables search for relevant records and displays a similarity metric between the query and the retrieved fragments, which, among other things, may serve as a recommendatory function for determining the causes of failures. Conclusion. The use of operation al inspection logs in combination with text mining methods can form the basis for building recommendation systems and decision support systems in the maintenance of railway infrastructure facilities.

Against the backdrop of the in-depth advancement of Digital China and the rural revitalization strategy, short video platform algorithms, as a novel cultural intermediary force, are intricately linked to the reconstruction of the political ecology of urban-rural cultural identity. Existing research on digital technology and rural development predominantly focuses on macro policy and micro individual behavior levels, lacking systematic empirical investigation into how platform algorithms, as a structural force, shape urban-rural cultural identity. This study employed a nationwide stratified sampling survey, with urban and rural residents as the research subjects, and utilized regression analysis and structural equation modeling to systematically examine the differential association mechanisms of algorithm recommendation systems on the cultural identity of urban and rural residents, as well as the moderating roles of social structural factors such as household registration and education level. The results revealed that algorithm exposure is significantly and positively correlated with users’ acceptance of rural modernity narratives, which is specifically reflected in the significant enhancement of fusion innovation identification. Urban-rural household registration, as a key social location variable, moderates the association path between algorithm exposure and reality identification: urban user groups exhibit a positive correlation between the two, whereas rural user groups show no such association. Active search behavior weakens the association with algorithm domestication, as users resist the infiltration of a single narrative through autonomous information acquisition. Notably, different short video platforms exhibit significant differences in their associations with cultural identity, and both the urbanization level of permanent residence and education level exert significant moderating effects on cultural identity and algorithm perception. Based on these findings, this study proposes the “Algorithm Domestication Gap” defining the digital cultural divide as a multi-dimensional cognitive gap within the framework of the third-generation digital divide. This concept extends the knowledge gap theory, providing a theoretical lens for understanding technology-mediated urban-rural cultural politics, and offers practical implications for digital rural construction and platform governance.

This study presents GeoCROP, a GIS-Based Crop Recommendation and Agricultural Data Management System developed to improve agricultural planning, data accessibility, and decision-making among farmers and agricultural stakeholders. Traditional agricultural practices often rely on fragmented and manual data handling, resulting in inefficiencies in monitoring crop distribution and accessing timely information. To address these challenges, GeoCROP was designed and developed using the Agile methodology, enabling iterative development and continuous system refinement. The system integrates Geographic Information System (GIS) technology to provide spatial visualization of farmer profiles, crop information, and agricultural locations. Core features include farmer profiling, crop management, GIS mapping, messaging, announcements, and report generation. The system was evaluated by IT experts and end-users using selected ISO/IEC 25010 quality criteria. Results indicate high technical quality, usability, and user acceptability, demonstrating that GeoCROP effectively supports agricultural data management and digital transformation initiatives.

Advances and challenges of multi-task learning method in recommender systems: A survey

Quantifying and mitigating the spiral of silence in recommender systems: A modular probabilistic framework

Trust in recommender systems: A survey

Recommender system for secure mobile application based on permission pairs using explainable artificial intelligence

Movie Recommendation system with sentiment analysis using deep learning algorithms

TLARS: A time and location-aware recommender system using dynamic multi-head attention

Improved DQN-based recommender system on three-way decision

Detecting shilling groups in recommender systems based on user multi-dimensional dynamic behavior analysis and graph contrastive learning

A knowledge graph enabled recommendation system for implicitly associated items: Application to vertical e-commerce of parts

Generative recommender systems: A comprehensive survey on model, framework, and application

This paper presents a novel modified seagull monarch butterfly optimization (MSMBO) algorithm, with a multi-objective focus on privacy and personalization in the fitness recommender system using a refined three-tier deep learning structure. The method is divided into three phases. In the first phase, fitness data from wearable devices undergoes preprocessing to eliminate noise and standardize features. The second phase incorporates improved elliptic curve cryptography (IECC) alongside the MSMBO to encrypt user data securely, ensuring privacy in cloud storage. This phase also enhances neural network performance by optimizing weights and hyperparameters through feature selection, effectively reducing data complexity while boosting accuracy. In the third phase, ConvCaps extracts spatial data features, while Bi-LSTM identifies temporal dependencies. The proposed system balances multiple objectives like novelty, accuracy, and precision, while safeguarding user data through robust encryption. With the experimental findings, our suggested method performs better than current existing models, especially in heart rate prediction and fitness pattern identification. The overall outcome makes the system ideal for privacyconscious, personalized fitness recommendations. The model’s shows significant improvement in mean squared error (MSE), normalized mean squared error (NMSE), and mean absolute percentage error (MAPE), thus verifying its effectiveness in secure, real-time fitness tracking.

Fuzzy-enhanced variable weight graph convolutional networks for recommender systems