Collaborative Filtering Recommendation with Fluctuations of User’ Preference

Yu, Jing; Shi, Jinaiing; Chen, Yunwen; Ji, Daqi; Liu, Wenhai; Xie, Zhijun; Li, Kai; Xue, Feng

doi:10.1109/icicse52190.2021.9404120

Cited by 24 publications

(6 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They refined recommendations using both collaborative and content-based filtering approaches, leveraging ratings from previous users to predict and recommend items to potential customers. [5] III.…”

Section: IImentioning

confidence: 99%

Movie Recommendation System Using Machine Learning

Parde,

Shirke,

Patil

et al. 2024

IJRASET

View full text Add to dashboard Cite

A recommendation engine utilizes various algorithms to filter information and suggest the most relevant items to users. It begins by analyzing a user's past behaviour and then recommends products based on that data. However, when a completely new user visits an e-commerce website, the site lacks any historical data on that user. In such cases, one approach could be to suggest popular products, those in high demand. Another approach might prioritize recommending products that would generate the most profit for the business. Three main approaches are commonly used in recommender systems: Demographic Filtering, Content-based Filtering, and Collaborative Filtering. Demographic Filtering provides generalized recommendations to users based on demographic features, such as movie quality or genre, assuming that users with similar demographics will have similar preferences. However, this approach is considered too simplistic since every user is unique. Content-based Filtering attempts to profile a user's interests using collected data and recommends items based on that profile. Collaborative Filtering groups similar users together and uses data about the group to make recommendations to individual users.

show abstract

Section: IImentioning

confidence: 99%

Movie Recommendation System Using Machine Learning

Parde,

Shirke,

Patil

et al. 2024

IJRASET

View full text Add to dashboard Cite

show abstract

“…Numerous studies and developments in personalized recom-mendation systems have been sparked by attempts to improve user experience in the online sphere [1]. A substantial amount of research explores the application of collaborative filtering methods to user preference gathering.…”

Section: Elated Workmentioning

confidence: 99%

“…Businesses prioritize an improved user experience as a means of connecting with their online audience in the rapidly changing digital market. This research project, which acknowledges the critical relevance of user pleasure, offers a novel solution based on the creation and application of an advanced computational framework [1]. The main goal is to produce extremely precise and user-focused product rankings that cater to the wide range of tastes of internet users.…”

Section: Introductionmentioning

confidence: 99%

Personalized product ranking system for enhanced user experience

Santhosh,

Ashweej,

Agarwal

et al. 2024

MATEC Web Conf.

View full text Add to dashboard Cite

Businesses looking to engage and satisfy their online audience must prioritize the user experience in the quickly changing digital landscape. In order to address this challenge, this project proposes and implements a sophisticated algorithmic solution designed to produce extremely accurate and user-centric product rankings. This represents a groundbreaking approach. The system attempts to predict and present the most relevant product suggestions by meticulously considering a wide range of factors such as user preferences, historical interactions, product popularity trends, and user similarity metrics. Our methodology is distinguished by the use of a dynamic simulation environment in which user profiles, product categories, and interaction pat- terns are manipulated to replicate authentic real-world scenarios. The dynamic framework facilitates the comprehensive testing and optimization of customized ranking algorithms, guaranteeing their flexibility in response to changing user preferences and behaviors. The project’s effectiveness is evaluated using precise evaluation metrics that offer quantitative information about how well the system understands and responds to each unique user’s preference, ultimately resulting in a more fulfilling and rich online shopping or content consumption experience.

show abstract

“…Zhang and Li (2010) argued that the user's interest is related to the item rated; if the user is more interested in an item, the decay rate decreases more slowly, while the user forgets the item more quickly if not interested in it. Yu et al (2021) added the factor of interest fluctuation to the traditional collaborative filtering system, quantifying interest fluctuation to recalculate predicted ratings. Joorabloo et al (2020) also considered changes in user interest preferences and proposed a method to modify neighborhood selection, considering the future similarity trends of users/items and reordering them accordingly.…”

Section: Introductionmentioning

confidence: 99%

UICF: a new user-item composite filtering recommendation framework by leveraging temporal semantics

Wei,

Liu,

Xian

et al. 2024

IJICC

View full text Add to dashboard Cite

PurposeThe collaborative filtering algorithm is a classical and widely used approach in product recommendation systems. However, the existing algorithms rely mostly on common ratings of items and do not consider temporal information about items or user interests. To solve this problem, this study proposes a new user-item composite filtering (UICF) recommendation framework by leveraging temporal semantics.Design/methodology/approachThe UICF framework fully utilizes the time information of item ratings for measuring the similarity of items and takes into account the short-term and long-term interest decay for computing users’ latest interest degrees. For an item to be probably recommended to a user, the interest degrees of the user on all the historically rated items are weighted by their similarities with the item to be recommended and then added up to predict the recommendation degree.FindingsComprehensive experiments on the MovieLens and KuaiRec datasets for user movie recommendation were conducted to evaluate the performance of the proposed UICF framework. Experimental results show that the UICF outperformed three well-known recommendation algorithms Item-Based Collaborative Filtering (IBCF), User-Based Collaborative Filtering (UBCF) and User-Popularity Composite Filtering (UPCF) in the root mean square error (RMSE), mean absolute error (MAE) and F1 metrics, especially yielding an average decrease of 11.9% in MAE.Originality/valueA UICF recommendation framework is proposed that combines a time-aware item similarity model and a time-wise user interest degree model. It overcomes the limitations of common rating items and utilizes temporal information in item ratings and user interests effectively, resulting in more accurate and personalized recommendations.

show abstract

Collaborative Filtering Recommendation with Fluctuations of User’ Preference

Cited by 24 publications

References 22 publications

Movie Recommendation System Using Machine Learning

Movie Recommendation System Using Machine Learning

Personalized product ranking system for enhanced user experience

UICF: a new user-item composite filtering recommendation framework by leveraging temporal semantics

Contact Info

Product

Resources

About