Movie Recommendation through Bias Analysis of Metadata

2021 ◽  
Vol 27 (10) ◽  
pp. 479-485
Author(s):  
Tae-Gyu Hwang ◽  
Sung Kwon Kim
Keyword(s):  
Bias Analysis ◽  
Movie Recommendation

scholarly journals Movie Recommendation through Multiple Bias Analysis

2021 ◽  
Vol 11 (6) ◽  
pp. 2817
Author(s):  
Tae-Gyu Hwang ◽  
Sung Kwon Kim
Keyword(s):  
Decision Making ◽  
Recommender System ◽  
Real World ◽  
Matrix Factorization ◽  
User Preferences ◽  
Improve Performance ◽  
Bias Analysis ◽  
Useful Knowledge ◽  
Existing Problems ◽  
Movie Recommendation

A recommender system (RS) refers to an agent that recommends items that are suitable for users, and it is implemented through collaborative filtering (CF). CF has a limitation in improving the accuracy of recommendations based on matrix factorization (MF). Therefore, a new method is required for analyzing preference patterns, which could not be derived by existing studies. This study aimed at solving the existing problems through bias analysis. By analyzing users’ and items’ biases of user preferences, the bias-based predictor (BBP) was developed and shown to outperform memory-based CF. In this paper, in order to enhance BBP, multiple bias analysis (MBA) was proposed to efficiently reflect the decision-making in real world. The experimental results using movie data revealed that MBA enhanced BBP accuracy, and that the hybrid models outperformed MF and SVD++. Based on this result, MBA is expected to improve performance when used as a system in related studies and provide useful knowledge in any areas that need features that can represent users.

Item Bias Analysis of the Young Schema-Questionnaire for Psychopathology, Gender, and Educational Level

2011 ◽  
Vol 27 (1) ◽  
pp. 65-70 ◽  
Author(s):  
Marleen M. Rijkeboer ◽  
Huub van den Bergh ◽  
Jan van den Bout
Keyword(s):  
Educational Level ◽  
Cross Validation ◽  
Item Bias ◽  
Response Patterns ◽  
Dutch Population ◽  
Bias Analysis ◽  
Young Schema Questionnaire ◽  
Item Functioning ◽  
Item Level ◽  
Schema Questionnaire

This study examines the construct validity of the Young Schema-Questionnaire at the item level in a Dutch population. Possible bias of items in relation to the presence or absence of psychopathology, gender, and educational level was analyzed, using a cross-validation design. None of the items of the YSQ exhibited differential item functioning (DIF) for gender, and only one item showed DIF for educational level. Furthermore, item bias analysis did not identify DIF for the presence or absence of psychopathology in as much as 195 of the 205 items comprising the YSQ. Ten items, however, spread over the questionnaire, were found to yield relatively inconsistent response patterns for patients and nonclinical participants.

2012 Workplace and Gender Relations Survey of Active Duty Members: Nonresponse Bias Analysis Report

2014 ◽  
Author(s):  
Fawzi Al-Nassir ◽  
Eric Falk ◽  
Owen Hung ◽  
Shoshana Magazine ◽  
Timothy Markheim ◽  
...  
Keyword(s):  
Gender Relations ◽  
Active Duty ◽  
Nonresponse Bias ◽  
Bias Analysis ◽  
And Gender

Effectiveness of Passion Fruit Peel Flour (Passiflora edulis L.) versus Turmeric Flour (Curcuma longa L.) on Glycemic Control: Systematic Review and Meta-Analysis

2020 ◽  
Vol 16 (5) ◽  
pp. 450-456
Author(s):  
Danilo F. Sousa ◽  
Vivian S. Veras ◽  
Vanessa E.C.S. Freire ◽  
Maria L. Paula ◽  
Maria A.A.O. Serra ◽  
...  
Keyword(s):  
Systematic Review ◽  
Glycemic Control ◽  
Curcuma Longa ◽  
Meta Analysis ◽  
Passion Fruit ◽  
Joint Analysis ◽  
Case Control Studies ◽  
Bias Analysis ◽  
Fruit Peel ◽  
Cross Sectional

Background:: It is undeniable that diabetes may cause several health complications for the population. Many of these complications are associated with poor glycemic control. Due to this, strategies to handle this problem are of great clinical importance and may contribute to reducing the various complications from diabetes. Objective: : The aim of this study was to compare the effectiveness of the passion fruit peel flour versus turmeric flour on glycemic control. Methods: This is a systematic review and meta-analysis following the PRISMA protocol. The following inclusion criteria were applied: (1) Case-control studies, cohort studies, and clinical trials, due to the improved statistical analysis and, in restrict cases, cross-sectional studies; (2) Articles published in any language. The databases used for the search were PubMed, Scopus, Web of Science, Cochrane, and LILACS. A bias analysis and a meta-analyses were undertaken using R Studio (version 3.3.1) using effect- size models. Results: : A total of 565 studies were identified from which 11 met the inclusion and exclusion criteria. Through isolated analysis, the effectiveness of turmeric flour on glycemic control was in the order of 0.73 CI (Confidence Interval) (from 0.68 to 0.79) and the effectiveness of passion fruit peel flour was 0.32 CI (0.23 to 0.45). The joint analysis resulted in 0.59 CI (0.52 to 0.68). The assessment of blood glucose was by glycated hemoglobin levels. All values were significant at a p < 0.05 level. Conclusion: Both interventions showed significant effects on glycemic control.

scholarly journals User profile correlation-based similarity (UPCSim) algorithm in movie recommendation system

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Triyanna Widiyaningtyas ◽  
Indriana Hidayah ◽  
Teguh B. Adji
Keyword(s):  
Collaborative Filtering ◽  
Recommendation System ◽  
User Behavior ◽  
Correlation Coefficients ◽  
User Profile ◽  
Profile Data ◽  
Similarity Algorithm ◽  
Previous Algorithm ◽  
Movie Recommendation ◽  
Recommendation Accuracy

AbstractCollaborative filtering is one of the most widely used recommendation system approaches. One issue in collaborative filtering is how to use a similarity algorithm to increase the accuracy of the recommendation system. Most recently, a similarity algorithm that combines the user rating value and the user behavior value has been proposed. The user behavior value is obtained from the user score probability in assessing the genre data. The problem with the algorithm is it only considers genre data for capturing user behavior value. Therefore, this study proposes a new similarity algorithm – so-called User Profile Correlation-based Similarity (UPCSim) – that examines the genre data and the user profile data, namely age, gender, occupation, and location. All the user profile data are used to find the weights of the similarities of user rating value and user behavior value. The weights of both similarities are obtained by calculating the correlation coefficients between the user profile data and the user rating or behavior values. An experiment shows that the UPCSim algorithm outperforms the previous algorithm on recommendation accuracy, reducing MAE by 1.64% and RMSE by 1.4%.

Sign in / Sign up

Export Citation Format

Share Document