Regret Bounds and Regimes of Optimality for User-User and Item-Item Collaborative Filtering

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.

Download Full-text

Comparative Analysis of Collaborative Filtering-Based Predictors of Scores in Surveys of a Large Company

10.5753/eniac.2021.18299 ◽

2021 ◽

Author(s):

Markos F. B. G. Oliveira ◽

Myriam Delgado ◽

Ricardo Lüders

Keyword(s):

Comparative Analysis ◽

Collaborative Filtering ◽

Matrix Factorization ◽

Relevant Alternatives ◽

Large Company ◽

Item Item

Collaborative Filtering (CF) can be understood as the process of predicting the preferences of users and deriving useful patterns by studying their activities. In the survey context, it can be used to predict answers to questions as combinations of other available answers. In this paper, we aim to test five CF-based algorithms (item-item, iterative matrix factorization, neural collaborative filtering, logistic matrix factorization, and an ensemble of them) to estimate scores in four survey applications (checkpoints) composed of 700,000 employee's ratings. These data have been collected from 2019 to 2020 by a large Brazilian tech company with more than 10,000 employees. The results show that collaborative filtering approaches provide relevant alternatives to score questions of surveys. They provided good quality estimates. This result can be further explored to eventually reduce the size of questionnaires, avoiding burden phenomena faced by respondents when dealing with large surveys.

Download Full-text

Group Recommender Systems: Combining User-User and Item-Item Collaborative Filtering Techniques

2015 International Conference on Information Technology (ICIT) ◽

10.1109/icit.2015.36 ◽

2015 ◽

Cited By ~ 3

Author(s):

Abinash Pujahari ◽

Vineet Padmanabhan

Keyword(s):

Collaborative Filtering ◽

Recommender Systems ◽

Group Recommender Systems ◽

Item Item ◽

Group Recommender ◽

Filtering Techniques

Download Full-text

Cloud based real-time collaborative filtering for item–item recommendations

Computers in Industry ◽

10.1016/j.compind.2013.11.005 ◽

2014 ◽

Vol 65 (2) ◽

pp. 279-290 ◽

Cited By ~ 8

Author(s):

Rafael Pereira ◽

Hélio Lopes ◽

Karin Breitman ◽

Vicente Mundim ◽

Wandenberg Peixoto

Keyword(s):

Collaborative Filtering ◽

Real Time ◽

Item Item

Download Full-text

Recommender system for surplus stock clearance

International Journal of Electrical and Computer Engineering (IJECE) ◽

10.11591/ijece.v9i5.pp3813-3821 ◽

2019 ◽

Vol 9 (5) ◽

pp. 3813

Author(s):

Vipul Agarwal ◽

Vijayalakshmi A

Keyword(s):

Collaborative Filtering ◽

Recommender Systems ◽

Recommender System ◽

Cosine Similarity ◽

Similarity Metrics ◽

Manual Work ◽

Similarity Metric ◽

Filtering Technique ◽

Retail Shop ◽

Item Item

Accumulation of the stock had been a major concern for retail shop owners. Surplus stock could be minimized if the system could continuously monitor the accumulated stock and recommend the stock which requires clearance. Recommender Systems computes the data, shadowing the manual work and give efficient recommendations to overcome stock accumulation, creating space for new stock for sale to enhance the profit in business. An intelligent recommender system was built that could work with the data and help the shop owners to overcome the issue of surplus stock in a remarkable way. An item-item collaborative filtering technique with Pearson similarity metric was used to draw the similarity between the items and accordingly give recommendations. The results obtained on the dataset highlighted the top-N items using the Pearson similarity and the Cosine similarity. The items having the highest rank had the highest accumulation and required attention to be cleared. The comparison is drawn for the precision and recall obtained by the similarity metrics used. The evaluation of the existing work was done using precision and recall, where the precision obtained was remarkable, while the recall has the scope of increment but in turn, it would reduce the value of precision. Thus, there lies a scope of reducing the stock accumulation with the help of a recommender system and overcome losses to maximize profit

Download Full-text

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

Mathematical Problems in Engineering ◽

10.1155/2014/162521 ◽

2014 ◽

Vol 2014 ◽

pp. 1-8 ◽

Cited By ~ 4

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.

Download Full-text