scholarly journals Advertising Popularity Feature Collaborative Recommendation Algorithm Based on Attention-LSTM Model

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Yang Su ◽  
Xiangwei Kong ◽  
Guobao Liu
Keyword(s):  
Prediction Model ◽  
Collaborative Filtering ◽  
Temporal Information ◽  
Experimental Results ◽  
Recommendation Algorithm ◽  
Filtering Method ◽  
Collaborative Recommendation ◽  
Percentage Points ◽  
Popularity Prediction ◽  
Click Through Rate

To accurately predict the click-through rate (CTR) and use it for ad recommendation, we propose a deep attention AD popularity prediction model (DAFCT) based on label recommendation technology and collaborative filtering method, which integrates content features and temporal information. First, we construct an Attention-LSTM model to capture the popularity trends and exploit the temporal information based on users’ feedback; finally, we use the concatenate method to fuse temporal information and content features and design a Deep Attention Popularity Prediction (DAVPP) algorithm to solve DAFCT. We experimentally adjust the weighted composite similarity metric parameters of Query pages and verify the scalability of the algorithm. Experimental results on the KDDCUP2012 dataset show that this model collaborative filtering and recommendation algorithm has better scalability and better recommendation quality. Compared with the Attention-LSTM model and the NFM model, the F1 score of DAFCT is improved by 9.80 and 3.07 percentage points, respectively.

A Collaborative Filtering Recommendation Algorithm Incorporated with Life Cycle

2013 ◽  
Vol 765-767 ◽  
pp. 630-633 ◽  
Author(s):  
Chong Lin Zheng ◽  
Kuang Rong Hao ◽  
Yong Sheng Ding
Keyword(s):  
Life Cycle ◽  
Collaborative Filtering ◽  
Recommendation Systems ◽  
Experimental Results ◽  
User Interest ◽  
Recommendation Algorithm ◽  
Time Information ◽  
Successful Technology

Collaborative filtering recommendation algorithm is the most successful technology for recommendation systems. However, traditional collaborative filtering recommendation algorithm does not consider the change of time information. For this problem,this paper improve the algorithm with two new methods:Predict score incorporated with time information in order to reflect the user interest change; Recommend according to scores by adding the weight information determined by the item life cycle. Experimental results show that the proposed algorithm outperforms the traditional item in accuracy.

scholarly journals Personalized Music Recommendation Simulation Based on Improved Collaborative Filtering Algorithm

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Hui Ning ◽  
Qian Li
Keyword(s):  
Collaborative Filtering ◽  
Recommendation System ◽  
Contextual Information ◽  
Rapid Development ◽  
Theoretical Research ◽  
Context Information ◽  
Music Recommendation ◽  
Recommendation Algorithm ◽  
Points Of Interest ◽  
Popularity Prediction

Collaborative filtering technology is currently the most successful and widely used technology in the recommendation system. It has achieved rapid development in theoretical research and practice. It selects information and similarity relationships based on the user’s history and collects others that are the same as the user’s hobbies. User’s evaluation information is to generate recommendations. The main research is the inadequate combination of context information and the mining of new points of interest in the context-aware recommendation process. On the basis of traditional recommendation technology, in view of the characteristics of the context information in music recommendation, a personalized and personalized music based on popularity prediction is proposed. Recommended algorithm is MRAPP (Media Recommendation Algorithm based on Popularity Prediction). The algorithm first analyzes the user’s contextual information under music recommendation and classifies and models the contextual information. The traditional content-based recommendation technology CB calculates the recommendation results and then, for the problem that content-based recommendation technology cannot recommend new points of interest for users, introduces the concept of popularity. First, we use the memory and forget function to reduce the score and then consider user attributes and product attributes to calculate similarity; secondly, we use logistic regression to train feature weights; finally, appropriate weights are used to combine user-based and item-based collaborative filtering recommendation results. Based on the above improvements, the improved collaborative filtering recommendation algorithm in this paper has greatly improved the prediction accuracy. Through theoretical proof and simulation experiments, the effectiveness of the MRAPP algorithm is demonstrated.

A Network-Based Multi-Dimensional Recommendation Algorithm

2013 ◽  
Vol 765-767 ◽  
pp. 1218-1222
Author(s):  
Xiang Yun Xiong ◽  
Yu Chen Fu ◽  
Zhao Qing Liu
Keyword(s):  
Collaborative Filtering ◽  
Experimental Results ◽  
Personalized Recommendation ◽  
Bipartite Network ◽  
Recommendation Algorithm ◽  
Better Than

Personalized recommendation based on bipartite network has attracted more and more attention. Its obviously better than CF (Collaborative Filtering). In this paper, we propose a multi-dimensional recommendation algorithm called BNPM. It combines item-based, user-based and category-based recommendation model to improve recommendation quality. The experimental results show that the algorithm can improve the diversity and reduce the popularity on the base of holding the accuracy of the recommendation

scholarly journals A Triangular Personalized Recommendation Algorithm for Improving Diversity

2018 ◽  
Vol 2018 ◽  
pp. 1-11 ◽  
Author(s):  
Biao Cai ◽  
Xiaowang Yang ◽  
Yusheng Huang ◽  
Hongjun Li ◽  
Qiang Sang
Keyword(s):  
Collaborative Filtering ◽  
Information Overload ◽  
Personalized Recommendation ◽  
Recommendation Algorithm ◽  
Filtering Method ◽  
Online Databases ◽  
Recommendation Algorithms ◽  
Recommendation Accuracy ◽  
Third Dimension ◽  
Triangular Area

Recommendation systems are used when searching online databases. As such they are very important tools because they provide users with predictions of the outcomes of different potential choices and help users to avoid information overload. They can be used on e-commerce websites and have attracted considerable attention in the scientific community. To date, many personalized recommendation algorithms have aimed to improve recommendation accuracy from the perspective of vertex similarities, such as collaborative filtering and mass diffusion. However, diversity is also an important evaluation index in the recommendation algorithm. In order to study both the accuracy and diversity of a recommendation algorithm at the same time, this study introduced a “third dimension” to the commonly used user/product two-dimensional recommendation, and a recommendation algorithm is proposed that is based on a triangular area (TR algorithm). The proposed algorithm combines the Markov chain and collaborative filtering method to make recommendations for users by building a triangle model, making use of the triangulated area. Additionally, recommendation algorithms based on a triangulated area are parameter-free and are more suitable for applications in real environments. Furthermore, the experimental results showed that the TR algorithm had better performance on diversity and novelty for real datasets of MovieLens-100K and MovieLens-1M than did the other benchmark methods.

Combining Item Rating Similarity and Item Classification Similarity for Better Recommendation Quality

2012 ◽  
Vol 461 ◽  
pp. 289-292
Author(s):  
Kai Zhou
Keyword(s):  
Collaborative Filtering ◽  
Recommender Systems ◽  
Fundamental Problem ◽  
The Other ◽  
Recommendation Algorithm ◽  
Data Sparsity ◽  
Filtering Method ◽  
Product Recommendation ◽  
Sparsity Problem

Recommender systems are becoming increasingly popular, and collaborative filtering method is one of the most important technologies in recommender systems. The ability of recommender systems to make correct predictions is fundamentally determined by the quality and fittingness of the collaborative filtering that implements them. It is currently mainly used for business purposes such as product recommendation. Collaborative filtering has two types. One is user based collaborative filtering using the similarity between users to predict and the other is item based collaborative filtering using the similarity between items. Although both of them are successfully applied in wide regions, they suffer from a fundamental problem of data sparsity. This paper gives a personalized collaborative filtering recommendation algorithm combining the item rating similarity and the item classification similarity. This method can alleviate the data sparsity problem in the recommender systems

A Collaborative Filtering Recommendation Algorithm Combined with User and Item

2014 ◽  
Vol 513-517 ◽  
pp. 1878-1881
Author(s):  
Feng Ming Liu ◽  
Hai Xia Li ◽  
Peng Dong
Keyword(s):  
Collaborative Filtering ◽  
Recommendation System ◽  
Experimental Results ◽  
Personalized Recommendation ◽  
User Evaluation ◽  
Recommendation Algorithm ◽  
Filtering Algorithm ◽  
Collaborative Filtering Algorithm

The collaborative filtering recommendation algorithm based on user is becoming the more personalized recommendation algorithm. But when the user evaluation for goods is very small and the user didnt evaluate the item, the commodity recommendation based on the item evaluation of user may not be accurate, and this is the sparseness in the collaborative filtering algorithm based on user. In order to solve this problem, this paper presents a collaborative filtering recommendation algorithm based on user and item. The experimental results show that this method has smaller MAE and greatly improve the quality of the recommendation in the recommendation system.

A Novel Personalized Filtering Recommendation Algorithm Based on Collaborative Tagging

2011 ◽  
Vol 186 ◽  
pp. 621-625
Author(s):  
Ming Yang Sun ◽  
Wei Feng Sun ◽  
Xi Dong Liu ◽  
Lei Xue
Keyword(s):  
Computational Complexity ◽  
Collaborative Filtering ◽  
Collaborative Tagging ◽  
Recommendation Algorithm ◽  
Trade Off ◽  
Filtering Method ◽  
Recommendation Algorithms ◽  
Sparsity Level ◽  
Personal Interests

Recommendation algorithms suffer the quality from the huge and sparse dataset. Memory-based collaborative filtering method has addressed the problem of sparsity by predicting unrated values. However, this method increases the computational complexity, sparsity and expensive complexity of computation are trade-off. In this paper, we propose a novel personalized filtering (PF) recommendation algorithm based on collaborative tagging, which weights the feature of tags that show latent personal interests and constructs a top-N tags set to filter out the undersized and dense dataset. The PF recommendation algorithm can track the changes of personal interests, which is an untilled field for previous studies. The results of empirical experiments show that the sparsity level of PF recommendation algorithm is much lower, and it is more computationally economic than previous algorithms.

Research of Data Sparsity Based on Collaborative Filtering Algorithm

2013 ◽  
Vol 462-463 ◽  
pp. 856-860
Author(s):  
Li Min Liu ◽  
Peng Xiang Zhang ◽  
Le Lin ◽  
Zhi Wei Xu
Keyword(s):  
Collaborative Filtering ◽  
Experimental Results ◽  
Recommendation Algorithm ◽  
Data Sparsity ◽  
Filtering Algorithm ◽  
Data Sparseness ◽  
Collaborative Filtering Algorithm ◽  
Scoring Matrix ◽  
Sparsity Problem ◽  
Improved Algorithm

During the traditional collaborative filtering recommendation algorithm be impacted by itself data sparseness problem. It can not provide accurate recommendation result. In this paper, Using traditional collaborative filtering algorithm and the concept of similar level, take advantage of the idea of data populating to solve sparsity problem, then using the Weighted Slope One algorithm to recommend calculating. Experimental results show that the improved algorithm solved the problem of the recommendation results of low accuracy because of the sparse scoring matrix, and it improved the algorithm recommended results to a certain extent.

Context-Aware Approach for Restaurant Recommender Systems

2019 ◽  
pp. 324-341 ◽  
Author(s):  
Haoxian Feng ◽  
Thomas Tran
Keyword(s):  
Root Mean Square Error ◽  
Collaborative Filtering ◽  
Recommender Systems ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Mean Square ◽  
Recommendation Algorithm ◽  
Filtering Method ◽  
User Behaviors ◽  
Svd Method

This chapter addresses the issue of how to effectively use users' historical data in restaurant recommender systems, as opposed to systems, such as FindMe, that only rely on online operations. Towards that end, the authors propose a bias-based SVD method as the underlying recommendation algorithm and test it against the traditional item-based collaborative filtering method on the Entrée restaurant dataset. The results are promising as the obtained root-mean-square-error (RMSE) values reach 0.58 for the SVD and 0.62 for the item-based system. Researchers can extend the transformation from user behaviors to ratings in more application domains other than the restaurant one.

Context-Aware Approach for Restaurant Recommender Systems

2018 ◽  
pp. 1757-1771
Author(s):  
Haoxian Feng ◽  
Thomas Tran
Keyword(s):  
Root Mean Square Error ◽  
Collaborative Filtering ◽  
Recommender Systems ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Mean Square ◽  
Recommendation Algorithm ◽  
Filtering Method ◽  
User Behaviors ◽  
Svd Method

This paper addresses the issue of how to effectively use users' historical data in restaurant recommender systems, as opposed to systems, such as FindMe, that only rely on online operations. Towards that end, the authors propose a bias-based SVD method as the underlying recommendation algorithm and test it against the traditional item-based collaborative filtering method on the Entrée restaurant dataset. The results are promising as the obtained Root-Mean-Square-Error (RMSE) values reach 0.58 for the SVD and 0.62 for the item-based system. Researchers can extend the transformation from user behaviors to ratings in more application domains other than the restaurant one.

Sign in / Sign up

Export Citation Format

Share Document