Clustering-Based Collaborative Filtering for Link Prediction

Wang, Xiangyu; He, Dayu; Chen, Danyang; Xu, Jinhui

doi:10.1609/aaai.v29i1.9162

Cited by 10 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cluster and process mining analysis were used to understand students' video engagements. Cluster analysis is a technique for grouping entities based on their common characteristics (Ungar & Foster, 1998). Based on the attributes of the entities, the algorithm divides the similarities among the individuals in the data set into a small number of sub-groups.…”

Section: Data Analysis Techniquesmentioning

confidence: 99%

Decoding Video Logs: Unveiling Student Engagement Patterns in Lecture Capture Videos

Akçapınar,

Er,

Bayazıt

2024

IRRODL

View full text Add to dashboard Cite

Lecture capture videos, a popular type of instructional content used by instructors to share course recordings online, play a significant role in educational settings. Compared to other educational videos, these recordings require minimal time and effort to produce, making them a preferred choice for disseminating course materials. Despite their numerous benefits, there exists a scarcity of data-driven evidence regarding students’ use of and engagement with lecture capture videos. Most existing studies rely on self-reported data, lacking comprehensive insights into students’ actual video engagement. This research endeavor sought to bridge this gap by investigating university students’ engagement patterns while watching lecture capture videos. To achieve this objective, we conducted an analysis of a large-scale dataset comprising over one million rows of video interaction logs. Leveraging clustering and process mining methodologies, we explored the data to reveal valuable insights into students’ video engagement behaviors. Our findings indicate that in approximately 60% of students’ video-watching sessions, only a small portion of the videos (an average of 7%) is watched. Our results also show that visiting the video page does not necessarily mean that the student watched it. This study may contribute to the existing literature by providing robust data-driven evidence on university students’ lecture capture video engagement patterns. It is also expected to contribute methodologically to capturing, preprocessing, and analyzing students’ video interactions in different contexts.

show abstract

Section: Data Analysis Techniquesmentioning

confidence: 99%

Decoding Video Logs: Unveiling Student Engagement Patterns in Lecture Capture Videos

Akçapınar,

Er,

Bayazıt

2024

IRRODL

View full text Add to dashboard Cite

show abstract

“…The authors compared the performance of the proposed re-clustering algorithm with the existing clustering algorithm and discussed their experimental result that their method improved the good quality of recommendation than others. Ungar and Foster (1998) have addressed the data sparsity problem by combining the clustering algorithm with singular value decomposition algorithm. In this method, they have used k-means clustering wherein the users are classified based on the similar attributes, and a new way of matrix representation called clustering-based rating matrix is formed.…”

Section: Related Workmentioning

confidence: 99%

Three-way formal concept clustering technique for matrix completion in recommender system

Selvi

Priya

2020

IJPCC

View full text Add to dashboard Cite

Purpose In today’s world, the recommender systems are very valuable systems for the online users, as the World Wide Web is loaded with plenty of available information causing the online users to spend more time and money. The recommender systems suggest some possible and relevant recommendation to the online users by applying the recommendation filtering techniques to the available source of information. The recommendation filtering techniques take the input data denoted as the matrix representation which is generally very sparse and high dimensional data in nature. Hence, the sparse data matrix is completed by filling the unknown or missing entries by using many matrix completion techniques. One of the most popular techniques used is the matrix factorization (MF) which aims to decompose the sparse data matrix into two new and small dimensional data matrix and whose dot product completes the matrix by filling the logical values. However, the MF technique failed to retain the loss of original information when it tried to decompose the matrix, and the error rate is relatively high which clearly shows the loss of such valuable information. Design/methodology/approach To alleviate the problem of data loss and data sparsity, the new algorithm from formal concept analysis (FCA), a mathematical model, is proposed for matrix completion which aims at filling the unknown or missing entries without loss of valuable information to a greater extent. The proposed matrix completion algorithm uses the clustering technique where the users who have commonly rated the items and have not commonly rated the items are captured into two classes. The matrix completion algorithm fills the mean cluster value of the unknown entries which well completes the matrix without actually decomposing the matrix. Findings The experiment was conducted on the available public data set, MovieLens, whose result shows the prediction error rate is minimal, and the comparison with the existing algorithms is also studied. Thus, the application of FCA in recommender systems proves minimum or no data loss and improvement in the prediction accuracy of rating score. Social implications The proposed matrix completion algorithm using FCA performs good recommendation which will be more useful for today’s online users in making decision with regard to the online purchasing of products. Originality/value This paper presents the new technique of matrix completion adopting the vital properties from FCA which is applied in the recommender systems. Hence, the proposed algorithm performs well when compared to other existing algorithms in terms of prediction accuracy.

show abstract

“…O'Connor and Herlocker [22] use clustering algorithms to partition the set of items, based on the user rating data. Ungar and Foster [28] combine separate clustering of users and items. The three above algorithms are all one-sided clustering, either for users or items.…”

Section: A Review Of User Clustering-based Rsmentioning

confidence: 99%

A new Recommendation Model for the User Clustering-Based Recommendation System

Rapečka

Dzemyda

2015

ITC

View full text Add to dashboard Cite

The aim of this paper is to create a new recommendation method that would evaluate the peculiarities of user groups, and to examine experimentally the efficiency of user clustering in order to improve the recommendations. To achieve this goal, we have analysed recommendation systems (RS), their components, operating principles and data, used for accuracy evaluation. The proposed method is based on user clustering; therefore, clustering-based RS are reviewed. Finally, the proposed method is presented and tested with the most appropriate data set of all that discussed in the overview. The research has disclosed dependencies of the efficiency of recommendations on the number of clusters. The experimental results have shown that the proposed method can be applied to high density databases and the results of recommendations are better than those of traditional methods.

show abstract

Clustering-Based Collaborative Filtering for Link Prediction

Cited by 10 publications

References 24 publications

Decoding Video Logs: Unveiling Student Engagement Patterns in Lecture Capture Videos

Decoding Video Logs: Unveiling Student Engagement Patterns in Lecture Capture Videos

Three-way formal concept clustering technique for matrix completion in recommender system

A new Recommendation Model for the User Clustering-Based Recommendation System

Contact Info

Product

Resources

About