scholarly journals A Soft-Rough Set Based Approach for Handling Contextual Sparsity in Context-Aware Video Recommender Systems

Mathematics ◽  
2019 ◽  
Vol 7 (8) ◽  
pp. 740 ◽  
Author(s):  
Syed Manzar Abbas ◽  
Khubaib Amjad Alam ◽  
Shahaboddin Shamshirband
Keyword(s):  
Recommender Systems ◽  
Rough Sets ◽  
Selection Process ◽  
Contextual Information ◽  
Context Aware ◽  
User Interactions ◽  
Information Selection ◽  
Proposed Model ◽  
Research Challenge ◽  
Selection Of

Context-aware video recommender systems (CAVRS) seek to improve recommendation performance by incorporating contextual features along with the conventional user-item ratings used by video recommender systems. In addition, the selection of influential and relevant contexts has a significant effect on the performance of CAVRS. However, it is not guaranteed that, under the same contextual scenario, all the items are evaluated by users for providing dense contextual ratings. This problem cause contextual sparsity in CAVRS because the influence of each contextual factor in traditional CAVRS assumes the weights of contexts homogeneously for each of the recommendations. Hence, the selection of influencing contexts with minimal conflicts is identified as a potential research challenge. This study aims at resolving the contextual sparsity problem to leverage user interactions at varying contexts with an item in CAVRS. This problem may be investigated by considering a formal approximation of contextual attributes. For the purpose of improving the accuracy of recommendation process, we have proposed a novel contextual information selection process using Soft-Rough Sets. The proposed model will select a minimal set of influencing contexts using a weights assign process by Soft-Rough sets. Moreover, the proposed algorithm has been extensively evaluated using “LDOS-CoMoDa” dataset, and the outcome signifies the accuracy of our approach in handling contextual sparsity by exploiting relevant contextual factors. The proposed model outperforms existing solutions by identifying relevant contexts efficiently based on certainty, strength, and relevancy for effective recommendations.

scholarly journals Context-Aware Neural Machine Translation for Korean Honorific Expressions

Electronics ◽  
2021 ◽  
Vol 10 (13) ◽  
pp. 1589
Author(s):  
Yongkeun Hwang ◽  
Yanghoon Kim ◽  
Kyomin Jung
Keyword(s):  
Machine Translation ◽  
Deep Neural Networks ◽  
Contextual Information ◽  
Context Aware ◽  
Neural Machine Translation ◽  
Translation Quality ◽  
Sentence Level ◽  
Proposed Model ◽  
The Given ◽  
The Relationship

Neural machine translation (NMT) is one of the text generation tasks which has achieved significant improvement with the rise of deep neural networks. However, language-specific problems such as handling the translation of honorifics received little attention. In this paper, we propose a context-aware NMT to promote translation improvements of Korean honorifics. By exploiting the information such as the relationship between speakers from the surrounding sentences, our proposed model effectively manages the use of honorific expressions. Specifically, we utilize a novel encoder architecture that can represent the contextual information of the given input sentences. Furthermore, a context-aware post-editing (CAPE) technique is adopted to refine a set of inconsistent sentence-level honorific translations. To demonstrate the efficacy of the proposed method, honorific-labeled test data is required. Thus, we also design a heuristic that labels Korean sentences to distinguish between honorific and non-honorific styles. Experimental results show that our proposed method outperforms sentence-level NMT baselines both in overall translation quality and honorific translations.

Transforming Geo-Referenced Data in Contextual Information for Context-Aware Recommender Systems

2018 ◽  
Author(s):  
Igor Andre Santana ◽  
Abner Suniga ◽  
Juliano Donini ◽  
Camila Vaccari Sundermann ◽  
Solange Oliveira Rezende ◽  
...  
Keyword(s):  
Recommender Systems ◽  
Contextual Information ◽  
Context Aware

A Hybrid Filtering Approach for an Improved Context-aware Recommender System

2019 ◽  
Vol 13 (1) ◽  
pp. 39-47 ◽  
Author(s):  
Mugdha Sharma ◽  
Laxmi Ahuja ◽  
Vinay Kumar
Keyword(s):  
Recommender Systems ◽  
Performance Improvement ◽  
Contextual Information ◽  
Hybrid Approach ◽  
Context Aware ◽  
Post Evaluation ◽  
Contextual Modeling ◽  
Filter Approach ◽  
Single Approach ◽  
Post Filter

Background: The domain of context-aware recommender approaches has made a substantial advancement over the last decade, but many applications still do not include contextual information while providing recommendations. Contextual information is crucial for various application areas and should not be ignored. Objective: There are generally three algorithms which can be used to include context and those are - pre-filter approach, post-filter approach and contextual modeling. Each of the algorithms has their own drawbacks if any single approach is chosen. The goal of this work is to identify and propose a new hybrid approach which can include contextual information to improve the current movie recommender systems. Method: Post evaluation of various patents related to recommender systems, the proposed approach modifies the post filter approach to rectify its shortcomings and combines it with the pre-filter approach based on the importance of contextual attribute provided by the user. Results: The performance of the proposed system is measured in terms of precision of the system and ranking of the recommended movies to the user. The results of experimental setup also demonstrate that the proposed system improves the precision and ranking of the recommendations provided to the user. Conclusion: With the help of this hybrid approach, the proposed system eliminates the problem of sparsity which is present in the pre-filter algorithm, and has performance improvement over the traditional post-filter approach. The proposed system will be vital for movie ticketing brands for the promotional purposes and various online content providers to recommend the accurate movies to their users.

Using Contextual Information from Topic Hierarchies to Improve Context-Aware Recommender Systems

2014 ◽  
Author(s):  
Marcos Aurelio Domingues ◽  
Marcelo Garcia Manzato ◽  
Ricardo Marcondes Marcacini ◽  
Camila Vaccari Sundermann ◽  
Solange Oliveira Rezende
Keyword(s):  
Recommender Systems ◽  
Contextual Information ◽  
Context Aware

Privileged contextual information for context-aware recommender systems

2016 ◽  
Vol 57 ◽  
pp. 139-158 ◽  
Author(s):  
Camila Vaccari Sundermann ◽  
Marcos Aurélio Domingues ◽  
Merley da Silva Conrado ◽  
Solange Oliveira Rezende
Keyword(s):  
Recommender Systems ◽  
Contextual Information ◽  
Context Aware

A multi-agent systems-based contractor pre-qualification model

2016 ◽  
Vol 23 (6) ◽  
pp. 709-726 ◽  
Author(s):  
Faikcan Kog ◽  
Hakan Yaman
Keyword(s):  
Selection Process ◽  
Construction Sector ◽  
Multi Agent Systems ◽  
Content Type ◽  
Quantitative Survey ◽  
Agent Systems ◽  
Proposed Model ◽  
Multi Agent ◽  
User Friendly ◽  
Selection Of

Purpose The selection of the contractor, as a main participant of a construction project, is the most important and challenging decision process for a client. The purpose of this paper is to propose a multi-agent systems (MAS)-based contractor pre-qualification (CP) model for the construction sector in the frame of the tender management system. Design/methodology/approach The meta-classification and analysis study of the existing literature on CP, contractor selection and criteria weighting issues, which examines the current and important CP criteria, other than price, is introduced structurally. A quantitative survey, which is carried out to estimate initial weightings of the identified criteria, is overviewed. MAS are used to model the pre-qualification process and workflows are shown in Petri nets formalism. A user-friendly prototype program is created in order to simulate the tendering process. In addition, a real case regarding the construction work in Turkey is analyzed. Findings There is a lack of non-human-driven solutions and automation in CP and in the selection problem. The proposed model simulates the pre-qualification process and provides consistent results. Research limitations/implications The meta-classification study consists of only peer-reviewed papers between 1992 and 2013 and the quantitative survey initiates the perspectives of the actors of Turkish construction sector. Only the traditional project delivery method is selected for the proposed model, that is other delivery methods such as design/build, project management, etc., are not considered. Open, selective limited and negotiated tendering processes are examined in the study and the direct supply is not considered in the scope. Practical implications The implications will help to provide an objective CP and selection process and to prevent the delays, costs and other troubles, which are caused by the false selection of a contractor. Originality/value Automation and simulation in the pre-qualification and the selection of the contractor with a non-human-driven intelligent solution ease the decision processes of clients in terms of cost, time and quality.

A State-of-the-Art Survey on Context-Aware Recommender Systems and Applications

2021 ◽  
Vol 12 (3) ◽  
pp. 1-20
Author(s):  
Quang-Hung Le ◽  
Son-Lam Vu ◽  
Thi-Kim-Phuong Nguyen ◽  
Thi-Xinh Le
Keyword(s):  
Recommender Systems ◽  
Recommendation System ◽  
Contextual Information ◽  
Information Filtering ◽  
Weather Conditions ◽  
Research Literature ◽  
Future Research ◽  
Context Aware ◽  
Good Tool ◽  
Advantages And Disadvantages

In the digital transformation era, increasingly more individuals and organizations use or create services in digital spaces. Many business transactions have been moving from the offline to online mode. For example, sellers intend to introduce their products on e-commerce platforms rather than display them on store shelves as in traditional business. Although this new format business has advantages, such as more space for product displays, more efficient searches for a specific item, and providing a good tool for both buyers and sellers to manage their products, it is also accompanied by the obviously important problem that users are confused when choosing an appropriate item due to a large amount of information. For this reason, the need for a recommendation system appears. Informally, a recommender system is similar to an information filtering system that helps identify a set of items that best satisfy users' demands based on their preference profiles. The integration of contextual information (e.g., location, weather conditions, and user's mood) into recommender systems to improve their performance has recently received considerable attention in the research literature. However, incorporating such contextual information into recommendation models is a challenging task because of the increase in both the dimensionality and sparsity of the model. Different approaches with their own advantages and disadvantages have been proposed. This paper provides a comprehensive survey on context-aware recommender systems in recent years. In particular, the authors pay more attention to journal and conference proceedings papers published from 2016 to 2020. In addition, this paper also presents open issues for context-aware recommender systems and discuss promising directions for future research.

Review Aware Recommender System

2018 ◽  
Vol 10 (2) ◽  
pp. 28-50
Author(s):  
Fatima Zahra Lahlou ◽  
Houda Benbrahim ◽  
Ismail Kassou
Keyword(s):  
Recommender Systems ◽  
Recommender System ◽  
State Of The Art ◽  
Contextual Information ◽  
Empirical Evaluation ◽  
Feature Engineering ◽  
Context Aware ◽  
Recommendation Accuracy

Context aware recommender systems (CARS) are recommender systems (RS) that provide recommendations according to user contexts. The first challenge for building such a system is to get the contextual information. Some works tried to get this information from reviews provided by users in addition to their ratings. However, all of these works perform important feature engineering in order to infer the context. In this article, the authors present a new CARS architecture that allows to automatically use contextual information from reviews without requiring any feature engineering. Moreover, they develop a new CARS algorithm that is tailored to textual contexts, that they call Textual Context Aware Factorization Machines (TCAFM). An empirical evaluation shows that the proposed architecture allows to significantly improve recommendation accuracy using state of the art RS and CARS algorithms, whereas TCAFM leads to additional improvements.

scholarly journals Privacy-preserving recommendations in context-aware mobile environments

2017 ◽  
Vol 25 (1) ◽  
pp. 62-79 ◽  
Author(s):  
Nikolaos Polatidis ◽  
Christos K. Georgiadis ◽  
Elias Pimenidis ◽  
Emmanouil Stiakakis
Keyword(s):  
Recommender Systems ◽  
Privacy Protection ◽  
Personal Information ◽  
Contextual Information ◽  
Context Aware ◽  
User Privacy ◽  
Data Set ◽  
Content Type ◽  
Privacy Concerns ◽  
Usability Tests

Purpose This paper aims to address privacy concerns that arise from the use of mobile recommender systems when processing contextual information relating to the user. Mobile recommender systems aim to solve the information overload problem by recommending products or services to users of Web services on mobile devices, such as smartphones or tablets, at any given point in time and in any possible location. They use recommendation methods, such as collaborative filtering or content-based filtering and use a considerable amount of contextual information to provide relevant recommendations. However, because of privacy concerns, users are not willing to provide the required personal information that would allow their views to be recorded and make these systems usable. Design/methodology/approach This work is focused on user privacy by providing a method for context privacy-preservation and privacy protection at user interface level. Thus, a set of algorithms that are part of the method has been designed with privacy protection in mind, which is done by using realistic dummy parameter creation. To demonstrate the applicability of the method, a relevant context-aware data set has been used to run performance and usability tests. Findings The proposed method has been experimentally evaluated using performance and usability evaluation tests and is shown that with a small decrease in terms of performance, user privacy can be protected. Originality/value This is a novel research paper that proposed a method for protecting the privacy of mobile recommender systems users when context parameters are used.

scholarly journals Deep Learning-Based Context-Aware Recommender System Considering Contextual Features

2021 ◽  
Vol 12 (1) ◽  
pp. 45
Author(s):  
Soo-Yeon Jeong ◽  
Young-Kuk Kim
Keyword(s):  
Deep Learning ◽  
Recommender System ◽  
Contextual Information ◽  
User Preferences ◽  
Context Aware ◽  
Data Sparsity ◽  
Contextual Features ◽  
Proposed Model ◽  
Multidimensional Matrix ◽  
Sparsity Problem

A context-aware recommender system can make recommendations to users by considering contextual information such as time and place, not only the scores assigned to items by users. However, as a user preferences matrix is expanded in a multidimensional matrix, data sparsity is maximized. In this paper, we propose a deep learning-based context-aware recommender system that considers the contextual features. Based on existing deep learning models, we combine a neural network and autoencoder to extract characteristics and predict scores in the process of restoring input data. The newly proposed model is able to easily reflect various type of contextual information and predicts user preferences by considering the feature of user, item and context. The experimental results confirm that the proposed method is mostly superior to the existing method in all datasets. Also, for the dataset with data sparsity problem, it was confirmed that the performance of the proposed method is higher than that of existing methods. The proposed method has higher precision by 0.01–0.05 than other recommender systems in a dataset with many context dimensions. And it showed good performance with a high precision of 0.03 to 0.09 in a small dimensional dataset.

Sign in / Sign up

Export Citation Format

Share Document