scholarly journals A Robust Collaborative Filtering Approach Based on User Relationships for Recommendation Systems

2014 ◽  
Vol 2014 ◽  
pp. 1-8 ◽  
Author(s):  
Min Gao ◽  
Bin Ling ◽  
Quan Yuan ◽  
Qingyu Xiong ◽  
Linda Yang
Keyword(s):  
Collaborative Filtering ◽  
Nearest Neighbor ◽  
Information Overload ◽  
Recommendation Systems ◽  
Personalized Recommendation ◽  
Dbscan Clustering ◽  
Shilling Attack ◽  
Typical Item ◽  
The Common ◽  
Item Item

Personalized recommendation systems have been widely used as an effective way to deal with information overload. The common approach in the systems, item-based collaborative filtering (CF), has been identified to be vulnerable to “Shilling” attack. To improve the robustness of item-based CF, the authors propose a novel CF approach based on the mostly used relationships between users. In the paper, three most commonly used relationships between users are analyzed and applied to construct several user models at first. The DBSCAN clustering is then utilized to select the valid user model in accordance with how the models benefit detecting spam users. The selected model is used to detect spam user group. Finally, a detection-based CF method is proposed for the calculation of item-item similarities and rating prediction, by setting different weights for suspicious spam users and normal users. The experimental results demonstrate that the proposed approach provides a better robustness than the typical item-basedkNN (kNearest Neighbor) CF approach.

scholarly journals Measuring the Interestingness of News Articles

2011 ◽  
pp. 1194-1199
Author(s):  
Raymond K. Pon ◽  
Alfonso F. Cardenas ◽  
David J. Buttler
Keyword(s):  
Collaborative Filtering ◽  
Information Overload ◽  
Recommendation Systems ◽  
Online News ◽  
Personalized Recommendation ◽  
Government Officials ◽  
Definition Of ◽  
News Recommendation ◽  
Keyword Searches ◽  
User Personalization

An explosive growth of online news has taken place. Users are inundated with thousands of news articles, only some of which are interesting. A system to filter out uninteresting articles would aid users that need to read and analyze many articles daily, such as financial analysts and government officials. The most obvious approach for reducing the amount of information overload is to learn keywords of interest for a user (Carreira et al., 2004). Although filtering articles based on keywords removes many irrelevant articles, there are still many uninteresting articles that are highly relevant to keyword searches. A relevant article may not be interesting for various reasons, such as the article’s age or if it discusses an event that the user has already read about in other articles. Although it has been shown that collaborative filtering can aid in personalized recommendation systems (Wang et al., 2006), a large number of users is needed. In a limited user environment, such as a small group of analysts monitoring news events, collaborative filtering would be ineffective. The definition of what makes an article interesting – or its “interestingness” – varies from user to user and is continually evolving, calling for adaptable user personalization. Furthermore, due to the nature of news, most articles are uninteresting since many are similar or report events outside the scope of an individual’s concerns. There has been much work in news recommendation systems, but none have yet addressed the question of what makes an article interesting.

Optimized Collaborative Filtering Recommendation Based on User’s Social Relationships

2014 ◽  
Vol 1044-1045 ◽  
pp. 1484-1488
Author(s):  
Yue Kun Fan ◽  
Xin Ye Li ◽  
Meng Meng Cao
Keyword(s):  
Collaborative Filtering ◽  
Social Relationships ◽  
Digital Libraries ◽  
Social Relationship ◽  
Nearest Neighbor ◽  
Mean Absolute Error ◽  
Absolute Error ◽  
Personalized Recommendation ◽  
Recommendation Algorithm

Currently collaborative filtering is widely used in e-commerce, digital libraries and other areas of personalized recommendation service system. Nearest-neighbor algorithm is the earliest proposed and the main collaborative filtering recommendation algorithm, but the data sparsity and cold-start problems seriously affect the recommendation quality. To solve these problems, A collaborative filtering recommendation algorithm based on users' social relationships is proposed. 0n the basis of traditional filtering recommendation technology, it combines with the interested objects of user's social relationship and takes the advantage of the tags to projects marked by users and their interested objects to improve the methods of recommendation. The experimental results of MAE ((Mean Absolute Error)) verify that this method can get better quality of recommendation.

Pre-Filling Collaborative Filtering Algorithm Based on Matrix Factorization

2013 ◽  
Vol 411-414 ◽  
pp. 2223-2228
Author(s):  
Dong Liang Su ◽  
Zhi Ming Cui ◽  
Jian Wu ◽  
Peng Peng Zhao
Keyword(s):  
Collaborative Filtering ◽  
Recommender System ◽  
Matrix Factorization ◽  
Nearest Neighbor ◽  
User Profile ◽  
Personalized Recommendation ◽  
System Quality ◽  
Recommendation Algorithm ◽  
Collaborative Filtering Algorithm ◽  
Personalized Recommender System

Nowadays personalized recommendation algorithm of e-commerce can hardly meet the needs of users as an ever-increasing number of users and items in personalized recommender system has brought about sparsity of user-item rating matrix and the emergence of more and more new users has threatened recommender system quality. This paper puts forward a pre-filled collaborative filtering recommendation algorithm based on matrix factorization, pre-filling user-item matrixes by matrix factorization and building nearest-neighbor models according to new user profile information, thus mitigating the influence of matrix sparsity and new users and improving the accuracy of recommender system. The experimental results suggest that this algorithm is more precise and effective than the traditional one under the condition of extremely sparse user-item rating matrix.

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.

An Ontology-Based Collaborative Filtering Personalized Recommendation

2012 ◽  
Vol 267 ◽  
pp. 79-82
Author(s):  
Pu Wang
Keyword(s):  
Collaborative Filtering ◽  
Web Sites ◽  
Information Overload ◽  
Cold Start ◽  
Personalized Recommendation ◽  
Collaborative Approach ◽  
The Matrix ◽  
Special Domain ◽  
Sparsity Problem ◽  
The Given

Recommender systems have been successfully used to tackle the problem of information overload, where users of products have too many choices and overwhelming amount of information about each choice. Personalization is widely used in various fields to provide users with more suitable and personalized service. Many e-commerce web sites such as online shop retailers make use of recommendation systems. In order to make recommendations to a user, collaborative filtering is an important personalized recommendation technique applied widely in E-commerce. The collaborative approach faces the hard issue of cold start problem and the matrix sparsity problem. The paper presents a collaborative filtering personalized recommendation approach based on ontology in the special domain. The method combines ontology technology and item-based collaborative filtering. The given recommendation approach can tackle the traditional recommenders problems, such as matrix sparsity and cold start problems.

scholarly journals A Hybrid Two-Phase Recommendation for Group-Buying E-commerce Applications

2019 ◽  
Vol 9 (15) ◽  
pp. 3141
Author(s):  
Li Bai ◽  
Mi Hu ◽  
Yunlong Ma ◽  
Min Liu
Keyword(s):  
Collaborative Filtering ◽  
Real World ◽  
Large Scale ◽  
Recommendation Systems ◽  
Personalized Recommendation ◽  
Two Phase ◽  
Group Buying ◽  
Large Scale Data ◽  
User Clustering ◽  
Scale Data

The last two decades have witnessed an explosive growth of e-commerce applications. Existing online recommendation systems for e-commerce applications, particularly group-buying applications, suffer from scalability and data sparsity problems when confronted with exponentially increasing large-scale data. This leads to a poor recommendation effect of traditional collaborative filtering (CF) methods in group-buying applications. In order to address this challenge, this paper proposes a hybrid two-phase recommendation (HTPR) method which consists of offline preparation and online recommendation, combining clustering and collaborative filtering techniques. The user-item category tendency matrix is constructed after clustering items, and then users are clustered to facilitate personalized recommendation where items are generated by collaborative filtering technology. In addition, a parallelized strategy was developed to optimize the recommendation process. Extensive experiments on a real-world dataset were conducted by comparing HTPR with other three recommendation methods: traditional CF, user-clustering based CF, and item-clustering based CF. The experimental results show that the proposed HTPR method is effective and can improve the accuracy of online recommendation systems for group-buying applications.

scholarly journals Design of Personalized Recommendation Algorithms based on Big Data Framework

2021 ◽  
Vol 2138 (1) ◽  
pp. 012025
Author(s):  
Fang Liu
Keyword(s):  
Big Data ◽  
Information Overload ◽  
Recommendation Systems ◽  
Recall Rate ◽  
Personalized Recommendation ◽  
Processing Technologies ◽  
Data Framework ◽  
Recommendation Algorithms ◽  
Data Environment ◽  
Better Than

Abstract The issue of information overload has become increasingly prominent since there are various kinds of data generated daily. A good recommendation systems can better deal with such problems. However, traditional recommendation systems for a single machine are suffering from the computing bottleneck in the environment of massive data. An individual recommendation algorithm is unable to gratify desiring users. To tackle this problem, we designed and implemented three kinds of recommendation algorithms based on big data framework in this paper. Besides, we improved the traditional recommendation algorithms leveraging the prevailing big data processing technologies. Finally, we evaluated the efficiency of the algorithm through recall rate, precision rate and coverage. Experiments show that the hybrid model-based recommendation algorithms which can be applied to the bulk data environment are better than the single recommendation algorithms.

scholarly journals Fusion of Internal Similarity to Improve the Accuracy of Recommendation Algorithm

2022 ◽  
Vol 24 (1) ◽  
pp. 139-140
Author(s):  
Dr.S. Dhanabal ◽  
◽  
Dr.K. Baskar ◽  
R. Premkumar ◽  
◽  
...  
Keyword(s):  
Collaborative Filtering ◽  
Recommender Systems ◽  
Information Overload ◽  
Real Life ◽  
Similarity Matrix ◽  
Recommendation Algorithm ◽  
Simulation Experiments ◽  
Average Similarity ◽  
Recommendation Accuracy ◽  
Item Item

Collaborative filtering algorithms (CF) and mass diffusion (MD) algorithms have been successfully applied to recommender systems for years and can solve the problem of information overload. However, both algorithms suffer from data sparsity, and both tend to recommend popular products, which have poor diversity and are not suitable for real life. In this paper, we propose a user internal similarity-based recommendation algorithm (UISRC). UISRC first calculates the item-item similarity matrix and calculates the average similarity between items purchased by each user as the user’s internal similarity. The internal similarity of users is combined to modify the recommendation score to make score predictions and suggestions. Simulation experiments on RYM and Last.FM datasets, the results show that UISRC can obtain better recommendation accuracy and a variety of recommendations than traditional CF and MD algorithms.

Research on Location-Based Personalized Recommendation System

2014 ◽  
Vol 490-491 ◽  
pp. 1493-1496
Author(s):  
Huan Gao ◽  
Xi Tian ◽  
Xiang Ling Fu
Keyword(s):  
Collaborative Filtering ◽  
Real Time ◽  
Recommendation System ◽  
Information Overload ◽  
Information Filtering ◽  
Mobile Internet ◽  
Personalized Recommendation ◽  
Recommendation Algorithm ◽  
New System

With the mobile Internet developing in China, the problem of information overload has been brought to us. The traditional personalized recommendation cannot meet the needs of the mobile Internet. In this paper, the recommendation algorithm is mainly based on the collaborative filtering, but the new factors are introduced into the recommendation system. The new system takes the user's location and friends recommendation into the personalized recommendation system so that the recommendation system can meet the mobile Internet requirements. Besides, this paper also puts forward the concept of moving business circle for information filtering, which realizes the precise and real-time personalized recommendations. This paper also proves the recommendation effects through collecting and analyzing the data, which comes from the website of dianping.com.

Personalized Recommendation Mechanism Based on Collaborative Filtering in Cloud Computing Environment

2021 ◽  
pp. 751-769
Author(s):  
Xinling Tang ◽  
Hongyan Xu ◽  
Yonghong Tan ◽  
Yanjun Gong
Keyword(s):  
Cloud Computing ◽  
Collaborative Filtering ◽  
Large Scale ◽  
Nearest Neighbor ◽  
Distributed Storage ◽  
Distributed Processing ◽  
Storage System ◽  
Personalized Recommendation ◽  
Computing Environment ◽  
Cloud Computing Environment

With the advent of cloud computing era and the dramatic increase in the amount of data applications, personalized recommendation technology is increasingly important. However, due to large scale and distributed processing architecture and other characteristics of cloud computing, the traditional recommendation techniques which are applied directly to the cloud computing environment will be faced with low recommendation precision, recommended delay, network overhead and other issues, leading to a sharp decline in performance recommendation. To solve these problems, the authors propose a personalized recommendation collaborative filtering mechanism RAC in the cloud computing environment. The first mechanism is to develop distributed score management strategy, by defining the candidate neighbors (CN) concept screening recommended greater impact on the results of the project set. And the authors build two stage index score based on distributed storage system, in order to ensure the recommended mechanism to locate the candidate neighbor. They propose collaborative filtering recommendation algorithm based on the candidate neighbor on this basis (CN-DCF). The target users are searched in candidate neighbors by the nearest neighbor k project score. And the target user's top-N recommendation sets are predicted. The results show that in the cloud computing environment RAC has a good recommendation accuracy and efficiency recommended.

Sign in / Sign up

Export Citation Format

Share Document