Improved Movie Recommendations Based on a Hybrid Feature Combination Method

Alshammari, Gharbi; Kapetanakis, Stelios; Alshammari, Abdullah; Polatidis, Nikolaos; Petridis, Miltos

doi:10.1142/s2196888819500192

Cited by 6 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A positive assortativity coefficient ( ) for a vertex attribute implies that vertices with that attribute have a high tendency to be connected. Formally, assortativity coefficient (ρ) for categorical attribute values is computed using Equation (12).…”

Section: 1homophily Of Communities Detected By Dsc Detectormentioning

confidence: 99%

Entropy for item inclination in sub-community based recommender system

2022

IJATEE

View full text Add to dashboard Cite

Machine learning (ML) entails self-learning via data usage and experience [1]. It requires no human intervention to uncover patterns in data [2]. Recent work in this area includes classification of liver tumours [3]; extraction of clinical attributes from breast cancer dataset [4]; prediction of student performance [5]; accurate recognition of complex physical human activity acquired using body-worn sensors [6]. Recommender system (RS) is also an exciting application of ML for suggesting relevant items to a user [7]. ML driven recommendation engines [8] have become ubiquitous in the last few decades. Intelligent web engines have crept in everywhere, recommending everything from movies, songs, food, social media posts to anything conceivable. Unconsciously, everybody is following these recommendations. The apparent reasons are convenience and satisfaction; else, dealing with profusion of information on the web is quite cumbersome. *Author for correspondence Famous online service providers like Facebook, Netflix, Spotify, Amazon, and LinkedIn use recommendation engines to boost sales and enhance customer satisfaction by utilising data filtering techniques of underlying RS [9]. RS uses traditional filtering techniques [10, 11] viz. collaborative filtering (CF), content-based filtering (CbF), demographic filtering (DF), and knowledgebased filtering (KF), along with hybrid filtering techniques that combine benefits of former techniques [12]. Liao et al. have found that users trust systems that use CF for recommendation over those using CbF or DF and have given pointers for solving cold-start problems [13]. CF is the most sought-after technique that collects users" preferences and predicts their interests. However, unfortunately it suffers from cold-start problems (non-availability of preference information for new user/item). Recent research points towards the inclusion of user demographic attributes (age, gender and location) to abate these problems [14]. Recently, González et al. have also utilized demographic information to evaluate the bias and unfairness of recommendations given to the minority groups [15].

show abstract

Section: 1homophily Of Communities Detected By Dsc Detectormentioning

confidence: 99%

Entropy for item inclination in sub-community based recommender system

2022

IJATEE

View full text Add to dashboard Cite

show abstract

“…Recommendation systems, which originated from the Internet and e-commerce industries, provide users with personalized information services and decision support based on data mining of magnanimity data [ 10 ]. It is an effective method to improve retrieval capabilities to help users obtain the required data, i.e., news recommendations, music recommendations, and movie recommendations [ 11 , 12 , 13 ]. With the increase of recommendation demand in various fields, scholars proposed many recommendation algorithms, such as content-based filtering (CBF), collaborative filtering (CF), and association-rule-based recommendation model (ARB) [ 14 , 15 , 16 ].…”

Section: Introductionmentioning

confidence: 99%

A Hybrid Recommendation System for Marine Science Observation Data Based on Content and Literature Filtering

Song

Guo

Chang

et al. 2020

Sensors

View full text Add to dashboard Cite

With the development of ocean exploration technology and the rapid growth in the amount of marine science observation data, people are faced with a great challenge to identify valuable data from the massive ocean observation data. A recommendation system is an effective method to improve retrieval capabilities to help users obtain valuable data. The two most popular recommendation algorithms are collaborative filtering algorithms and content-based filtering algorithms, which may not work well for marine science observation data given the complexity of data attributes and lack of user information. In this study, an approach was proposed based on data similarity and data correlation. Data similarity was calculated by analyzing the subject, source, spatial, and temporal attributes to obtain the recommendation list. Then, data correlation was calculated based on the literature on marine science data and ranking of the recommendation list to obtain the re-rank recommendation list. The approach was tested by simulated datasets collected from multiple marine data sharing websites, and the result suggested that the proposed method exhibits better effectiveness.

show abstract