From Theory to Practice: The Evolution and Comparative Analysis of Homogeneous vs. Heterogeneous Graph Neural Networks in Recommender Systems

Neurocomputing, Год журнала: 2025, Номер unknown, С. 129446 - 129446

Опубликована: Янв. 1, 2025

Язык: Английский

---

Research on the Problems of Teaching Traditional Music Basic Theory in the Information Age

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract In the era of informationization, inheritance and development traditional music are facing challenges opportunities. This study analyzes problems faced by teaching basic theory proposes a coping strategy for constructing an interactive model platform informatization. model, P2P technology is used to design construct behavior interest analysis module, fuzzy precisely set key information weights. The informationized designed with AI technology, adopting recommendation hybrid combined collaborative filtering, using cosine similarity calculate content similarity, combining artificial intelligence realize quality assessment. practice, S High School as practice site, experimental control classes up conduct teaching. class perceived negative behaviors such “harshness” “dissatisfaction” lower than class, pass rate excellence post-test reached 92.5% 36.9%, respectively. mean values all other core literacy dimensions except knowledge opera general musical instruments were higher those showing significant differences (P<0.05).

Язык: Английский

---

Human resource management and artificial intelligence integration development and innovation

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract Artificial Intelligence is one of the important tools to realize effective management human resources. The article firstly investigates and finds out problems current salary satisfaction status employees column A a news center, designs job prediction model based on collaborative filtering algorithm separation standard Logistic regression results questionnaire, finally calculates economic benefits center. performance bonus center in 2019 showed steady upward trend as whole, with 107,908 yuan January 198,320 December. market value increased steadily 2020, reaching 171,134 230,370 In 2021, ~ June has not changed much, which 112,099 122,076 June. It can be concluded that proposed this paper high accuracy predicting column, shows an increasing year by year. seen from analysis after operating financial have been for better.

Язык: Английский

---

Optimizing Collaborative Filtering Recommendation Algorithms for Knowledge Sharing in Libraries

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract With the rapid development of information technology, how to enable teachers and students quickly find filter interest in massive collections has become a hot issue for many scholars. This paper enhances traditional collaborative filtering algorithm by utilizing knowledge-sharing model. Specifically, we calculate similarity using keyword from items dynamic users based on calculations related item features user attributes. The relevant about is fully utilized successfully alleviate problem new users, entire process recommendations libraries designed. improved filtering-based can achieve recommendation accuracy more than 60% recommend accurate books users. average rate book 0.056 higher other recommender systems, indicating that better match needs cold-start certain extent.

Язык: Английский

---

Artificial Intelligence Technology Facilitates the Design of Interactive Platform for English Education in Colleges and Universities

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract In today’s era of rapid scientific and technological advancements, the deep integration artificial intelligence technology education is paving way for intelligent development in English education. This paper designs architecture interactive platform colleges universities based on technology, provides relevant tools interaction assistance, establishes an analysis model students’ online learning interactions. To promote diversity test papers, improved genetic algorithm utilised grouping papers. A collaborative filtering used to intelligently predict grades by considering time sequence courses. Performance tests were designed verify effectiveness application The took average 1.247 seconds, discrepancy between predicted actual course did not surpass one grade. had a minimum response 1.69 ms, server didn’t use more than 10% its total CPU. platform, which combines with education, new path innovating teaching methods

Язык: Английский

---

Research on Recommendation Methods for Scientific and Technological Information and Their Application in College Education - Based on Knowledge Graphs

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract In the current technology information recommendation system, construction of user behavior matrix frequently results in sparsity and cold start issues. To address this problem, TransAR-CF, a knowledge graph-based method, is proposed, which mainly combines item similarity semantic to form hybrid then accomplishes task constructing model. The model applied college education by service system experience are tested separately. convergence speed paper’s higher than that CFKG on both datasets A B, difference between two ranges from 0.02 0.12. According four dimensions design, learning support, evaluation teaching effect, 70% 90% users have positive attitude toward system. summary, TransARCF method solves problem scientific technological difficult adapt accurately but has high acceptance good utility education.

Язык: Английский

---

Innovation of Teaching Methods of Civic and Political Education in Colleges and Universities with the Assistance of Artificial Intelligence

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract Artificial intelligence empowers education and promotes its integration innovation, enabling the use of learning analytics to accurately diagnose situation implement precise teaching. In this paper, we have leveraged artificial innovate traditional teaching methods, developed a wisdom model specifically for college civic education, concentrated on application in context. First, it creates student profiles based academic data then provides personalized recommendations resources by combining collaborative filtering algorithm an improved K-means algorithm. Finally, analyze actual impact intervention comparing portraits before after intervention. When number neighbors K takes same value, cosine similarity calculation method outperforms other RMSE, MAE, accuracy, recall, F1 values are 0.14, 0.24, 0.99, 0.74, 0.85 when = 60, respectively, which indicates that improvement paper is practical. This categorizes sample students into four groups their behavior outcomes: potential, excellent, marginal, hardworking. All these types demonstrated some progress following intervention, confirming efficacy paper’s intelligence-based approach education. Simultaneously, created establish groundwork future accurate instruction.

Язык: Английский

---

Exploring the Potential Application of Big Data in the Digital Transformation Practice of English Education

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract With the increasing development of digital economy and Internet technology, in order to adapt demand for teaching, many colleges universities are changing their education mode through transformation. Starting from English this paper seeks uncover a suitable way transformation education. Based on collaborative filtering recommendation algorithm, constructs an teaching model that integrates personality recommendations. The can use clustering algorithms classify learners into different groups, which facilitates subsequent implementation personalized performance comparison examples used analyze efficient evaluate its usefulness smart classroom. designed divides five clusters, provides strong support us provide recommendations learning resources future. This paper’s system, example, basic English, has accuracy rate 10% 21% higher than reference systems 1 2, according comparison. Applying intelligent classroom, after experiment, class T improved by 4.29 points compared CK. To conclude, system is very adaptable classroom greater role promoting universities.

Язык: Английский

---

Study on the effect of dissemination of Luo and Yue rice culture in multimedia technology-assisted rural cultural revitalization

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract Algorithmic recommendation is an important means of using multimedia technology for cultural dissemination and helping rural revitalization. In order to realize the effective Luo Yue rice culture, this paper constructs a hybrid algorithm that recommends videos about culture according user preferences, which uses content relationship derived from LDA replace list generated by collaborative filtering crosses it with historical rating solve problems accuracy cold start. The analysis effect shows average MAE value paper’s 0.671, proportion users who watch relevant “occasionally” more frequently has increased 5.84%, recognize method. proposed in good performance helps disseminate as demonstrated above results.

Язык: Английский

---

Research on the Design of Intelligent Recommendation System for Civic and Political Education Content and the Effectiveness Assessment of Students’ Acceptance

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract The article improves the traditional collaborative filtering algorithm, integrates it with content-based recommendation proposes algorithm based on mixture of and content, serves as operation logic for designing intelligent system educational content Civics class. Model variables are determined using structural equation modeling relevant hypotheses presented to construct a model factors that influence student acceptance Intelligent Recommendation System, followed by empirical analysis. mean values expectation performance, effort social influence, convenience conditions, self-efficacy, perceived pleasantness, willingness use 3.48, 2.70, 3.61, 2.36, 3.77, 3.84, 3.73, respectively. Students’ Civic System is greatly influenced their perception pleasantness self-efficacy. questionnaire has good reliability validity in general. initial valid H1, H2, H6, H7, H8, H9, H10, H11. In analysis variance, there were significant differences between genders only performance expectations (0.000) (0.016). Significant existed across grades terms (0.018) (0.000). measurement dimension had moderating effect majors. Hypotheses H12 H13 partially valid, but H14 valid.

Язык: Английский

---