Improved Recommendation Algorithm Research Articles

Personalized Recommendation System for English Teaching Resources in Colleges and Universities Based on Collaborative Recommendation

Abstract In recent years, with the rapid development of the Internet and education informatization, online teaching has become a popular education mode in the information age, providing learners with very rich teaching resources. In this paper, we construct a personalized system and co-recommendation technology for English teaching resources, and we improve the traditional co-recommendation algorithm and propose hybrid recommendations. The performance of the system is evaluated experimentally to compare the effectiveness of the performance of the four systems. The improved recommendation algorithm is superior to the other three recommendation algorithms in each dataset in the system. The average grade mean of the experimental class assisted by the hybrid recommendation system in teaching English in colleges and universities in the latter two experiments is 27.54, which is higher than that of the comparison class of 25.33, and the T-value is 1.81>1.645. The improved personalized recommender system has good validity and stability in both performance and practical application.

The Application of Digital Multimedia Technology in the Innovative Mode of English Teaching in institutions of higher education

Abstract Innovative English teaching resources will inevitably result from the use of media technology in the classroom. This article will examine the new approach to teaching English in college from the standpoint of teaching resources. In order to establish a new mode of English teaching in higher education institutions based on personalized recommendation of teaching resources, this paper proposes two recommendation algorithms: one is content-based, and the other is collaborative filtering. The algorithm is enhanced by combining it with the design of multimedia English teaching resources. The influence of hit rate and MAE values prior to and following the cooperative filtering recommendation method is examined. The distinction between the pre-and post-test results of the experimental and control courses is analyzed and compared to conclude the examination of the impact of the novel model of teaching college English. The analysis indicates that the improved recommendation algorithm has a 4.85% higher hit rate than the traditional collaborative filtering algorithm. The improved recommendation algorithm is more accurate than the traditional recommendation algorithm, as suggested by this.

The Recommendation Algorithm Based on Improved Conditional Variational Autoencoder and Constrained Probabilistic Matrix Factorization

An improved recommendation algorithm based on Conditional Variational Autoencoder (CVAE) and Constrained Probabilistic Matrix Factorization (CPMF) is proposed to address the issues of poor recommendation performance in traditional user-based collaborative filtering algorithms caused by data sparsity and suboptimal feature extraction. Firstly, in the data preprocessing stage, a hidden layer is added to CVAE, and random noise is introduced into the hidden layer to constrain the data features, thereby obtaining more accurate latent features and improving the model’s robustness and generative capability. Secondly, the category of items is incorporated as auxiliary information in CVAE to supervise the encoding and decoding of item data. By learning the distribution characteristics of the data, missing values in the rating data can be effectively reconstructed, thereby reducing the sparsity of the rating matrix. Subsequently, the reconstructed data is processed using CPMF, which optimizes the feature extraction performance by imposing constraints on user features. Finally, the prediction rating of a user for an item can be obtained through the matrix product of user and item feature matrices. Experimental results on the MovieLens-100K and MovieLens-1M datasets demonstrate the effectiveness and superiority of the proposed algorithm over four comparative algorithms, as it exhibits significant advantages in terms of root mean square error and mean absolute error metrics.

Study on the Application of Ganan Hakka Architectural Elements in Modern Architectural Interior Design

Abstract This paper constructs a morphological extraction model of Ganan Hakka architectural elements based on feature fusion using ResNet as the backbone network. Then, the extraction method of Ganan Hakka architectural colors based on the combination of multivariate fuzzy C-mean clustering algorithm and the K-mean++ algorithm is designed. The improved recommendation algorithm intelligently matches the extracted color and morphological elements and applies them to modern architectural interior design. The results show that the average time of interior design intelligent recommendation is 3.8s, the overall accuracy is above 0.94, and the degree of interior design intelligent matching is in the range of [0.78,0.99]. Based on this study, Ganan Hakka’s architectural elements can be well combined with modern architectural interiors to promote traditional culture.

The Construction of Cultural Translation Research System under Big Data Technology

Abstract In the context of big data development, one of the reasons for the ineffectiveness of Chinese cultural translation research is the lack of attention to the “system construction”, this paper focuses on the construction of the cultural translation research system under the big data technology, based on the MOOC recommendation system using the recommendation engine group for the hybrid recommendation, the recommendation engine group integrated the use of the mainstream collaborative filtering algorithm based on foreign translation literature in the recommendation algorithm. Through the experimental analysis, the model performance reaches the maximum when the number of cultural literature is 7, and the F_1 of the traditional method is improved by about 2.6%. Compared with the original recommendation algorithm, the improved recommendation algorithm of the MOOC implicit scoring model can get more accurate recommendation results to promote the construction of a Chinese cultural foreign translation discourse system. It is expected to promote the construction of Chinese cultural foreign translation discourse system, show the vitality, staying power, and influence of China’s excellent culture to the world, enhance the international discourse power and cultural soft power of Chinese culture in the new era, and contribute Chinese wisdom to global cultural exchange and governance.

Simulation of book selection planning based on deep learning and its application

Due to the complexity of network information, it presents an unbalanced feature. For the current book publishers, the publishing theme selection and planning methods of the newsroom can no longer meet the development speed of the Internet background. As the publishing house cannot accurately analyze the needs of users, it is difficult to obtain the specific standards of the book publishing market. Therefore, the demand of consumers for books has decreased, and further generated practical problems such as inventory accumulation and revenue impairment. Based on the demand of extracting and analyzing target information for books, it can be realized by using deep learning methods. Therefore, this study establishes a book selection planning system. If the information of the book itself and the corresponding evaluation information are required, the system first uses Anaconda to crawl the website to obtain the book information data, and then uses the proposed KIEM algorithm to supplement the information of the data crawled by Anaconda. After completing this step, use the first layer of CRF model to separate the evaluation sentence from the opinion sentence, and split the separated target sentence to get the attribute and emotion words. Finally, ARIA algorithm is combined with the improved recommendation algorithm to improve the recommendation performance and achieve the accuracy and personalization of the system. It can be seen from the actual measurement that the MAE value calculated by the application of the system proposed in this paper is relatively small, which shows that the system has certain practicability in prediction accuracy. This paper designs an effective system application by introducing deep learning technology into the field of book topic selection planning.

Development and Replication of Improved Recommendation Algorithm for Network Representation Learning

Development and Replication of Improved Recommendation Algorithm for Network Representation Learning

An Improved Recommendation Algorithm For Polarized Population

An Improved Recommendation Algorithm For Polarized Population

An Improved Recommendation Algorithm Based on Two-layer Attention Mechanism

An Improved Recommendation Algorithm Based on Two-layer Attention Mechanism

Product advertising recommendation in e-commerce based on deep learning and distributed expression

With the advent of Internet big data era, recommendation system has become a hot research topic of information selection. This paper studies the application of deep learning and distributed expression technology in e-commerce product advertising recommendation. In this paper, firstly, from the semantic level of advertising, we build a similarity network based on the theme distribution of advertising, and then build a deep learning model framework for advertising click through rate prediction. Finally, we propose an improved recommendation algorithm based on recurrent neural network and distributed expression. Aiming at the particularity of the recommendation algorithm, this paper improves the traditional recurrent neural network, and introduces a time window to control the hidden layer data transfer of the recurrent neural network. The experimental results show that the improved recurrent neural network model based on time window is superior to the traditional recurrent neural network model in the accuracy of recommendation system. The complexity of calculation is reduced and the accuracy of recommendation system is improved.

Research on Hybrid Recommendation Model Based on PersonRank Algorithm and TensorFlow Platform

At present, the borrowing data of university libraries in western China generally has the problem of large sparseness and inaccurate recommendation results. However, the traditional recommendation algorithm does not solve the problem. Therefore, solving the data sparsity problem and improving the recommendation accuracy has become a very important issue. Through research, the author proposes a hybrid recommendation model that combines the PersonRank algorithm with the neural network trained by Google’s artificial intelligence framework TensorFlow in the context of excessive data sparseness in western college libraries. First, using the existing borrowing records to model through the bipartite graph probability random walk method, the author obtains a list of books that the reader may be interested in and calculates the probability value of interest. After that, the author converts the probability value into a pseudo-scoring table, and uses TensorFlow to calculate by pseudo-scoring, and supplements the data training set and book classification information to obtain the recommendation result. The model effectively solves the problem of data sparsity in the recommendation system. Finally, it compares with the traditional recommendation algorithm and an improved recommendation algorithm. The accuracy of the model is evaluated by AUC index, and the accuracy of the proposed model is higher.

On the Risk Assessment of Terrorist Attacks Coupled with Multi-Source Factors

Terrorism has wreaked havoc on today’s society and people. The discovery of the regularity of terrorist attacks is of great significance to the global counterterrorism strategy. In this study, we improve the traditional location recommendation algorithm coupled with multi-source factors and spatial characteristics. We used the data of terrorist attacks in Southeast Asia from 1970 to 2016, and comprehensively considered 17 influencing factors, including socioeconomic and natural resource factors. The improved recommendation algorithm is used to build a spatial risk assessment model of terrorist attacks, and the effectiveness is tested. The model trained in this study is tested with precision, recall, and F-Measure. The results show that, when the threshold is 0.4, the precision is as high as 88%, and the F-Measure is the highest. We assess the spatial risk of the terrorist attacks in Southeast Asia through experiments. It can be seen that the southernmost part of the Indochina peninsula and the Philippines are high-risk areas and that the medium-risk and high-risk areas are mainly distributed in the coastal areas. Therefore, future anti-terrorism measures should pay more attention to these areas.

An improved recommendation algorithm for big data cloud service based on the trust in sociology

Personal recommendation technology is becoming a useful and popular solution to solve the problem of information overload with the popularity of big data cloud services. But most recommendation algorithms pay too much attention to the similarity to focus on the social trust between users. So this paper focus on the research of hybrid Recommendation algorithm for big data based on the optimization combining with the similarity and trust in sociology. In this paper, we introduced some user trust models including trust path model and loop trust model, and then we took these models into the calculation of mixed weighting. The experiment results show that the recommendation algorithm considering the trust models has the higher accuracy than the traditional recommendation algorithm, and we have a 2% increase in both MEA (Mean Absolute Error) and RMSE (Root Mean Square Error).

An improved recommendation algorithm via weakening indirect linkage effect**Project supported by the National Natural Science Foundation of China (Grant No. 11175079) and the Young Scientist Training Project of Jiangxi Province, China (Grant No. 20133BCB23017).

We propose an indirect-link-weakened mass diffusion method (IMD), by considering the indirect linkage and the source object heterogeneity effect in the mass diffusion (MD) recommendation method. Experimental results on the MovieLens, Netflix, and RYM datasets show that, the IMD method greatly improves both the recommendation accuracy and diversity, compared with a heterogeneity-weakened MD method (HMD), which only considers the source object heterogeneity. Moreover, the recommendation accuracy of the cold objects is also better elevated in the IMD than the HMD method. It suggests that eliminating the redundancy induced by the indirect linkages could have a prominent effect on the recommendation efficiency in the MD method.

Improved Collaborative Filtering Recommendation Algorithm Based on Weighted Association Rules

In this paper the collaborative filtering recommendation and association rules are introduced firstly. Aiming at the shortcomings of basic collaborative filtering recommendation, this paper proposes an improved collaborative filtering recommendation algorithm based on weighted association rules (CFRA-WAR). Finally the simulation experiments are carried out to verify the validity of the improved recommendation algorithm. The results of simulation experiments show that the recommendation accuracy of CFRA-WAR is superior to basic collaborative filtering recommendation algorithm, although the algorithm time of CFRA-WAR is a little longer.

An Improved Recommendation Algorithm Based on Graph Model

According to the problem that the traditional search algorithms dont consider the needs of individuals, various recommender systems employing different data representations and recommendation methods are currently used to cope with these challenges. In this paper, inspired by the network-based user-item rating matrix, we introduce an improved algorithm which combines the similarity of items with a dynamic resource allocation process. To demonstrate its accuracy and usefulness, this paper compares the proposed algorithm with collaborative filtering algorithm using data from MovieLens. The evaluation shows that, the improved recommendation algorithm based on graph model achieves more accurate predictions and more reasonable recommendation than collaborative filtering algorithm or the basic graph model algorithm does.

An Improved Recommendation Algorithm in Knowledge Network

Knowledge sharing is an essential characteristic of knowledge network and knowledge recommendation is a good way to improve the quality of shared knowledge. To investigate more efficient knowledge recommendation algorithm, knowledge potential energy rank and interest measure of knowledge were added in the association rules. Augmentation and quantization of the two parameters is the most important innovation of the paper. There were three modules in the algorithm-mining knowledge association rules, recommending related knowledge and modifying knowledge rank. Examination process provided by coal-wz.com was used to verify the proposed approach. The results showed that the proposed approach not only is able to extract the rules more efficient and much faster, but also can discover association rules more accurate in the context of knowledge management in knowledge network.

Weighted bipartite network and personalized recommendation

In this paper, the degree distributions of a bipartite network, namely Movielens, are investigated. The statistical analysis shows that the distribution of the degree product, ku ko, has an exponential from, where ku and ko denote the user and object degrees respectively. By introducing the edge weight effect on the recommendation performance, an improved recommendation algorithm based on mass diffusion (MD) process is presented. We argue that the edges weight of the user-object bipartite network should be taken into account to measure the object similarity. By taking into account the user and object degree correlations, the weighted bipartite network is constructed. The numerical results of the MD algorithms on the weighted network indicate that both of the accuracy and diversity could be increased at the optimal case. More importantly, we find that, at the optimal case, the edge weight distribution would change from the exponential form to the poisson form. This work may shed some light on how to improve the recommendation algorithm performance by considering the statistical properties.