Multi-Agent-Based Recommender Systems: A Literature Review

Trust-Aware Recommender Systems for Open and Mobile Virtual Communities

Information Resources Management ◽

10.4018/978-1-61520-965-1.ch210 ◽

2010 ◽

pp. 326-345

Author(s):

N. Sahli ◽

G. Lenzini

Keyword(s):

Recommender Systems ◽

Virtual Communities ◽

Multi Agent Systems ◽

Artificial Intelligent ◽

Agent Based ◽

Agent Systems ◽

Multi Agent

This chapter surveys and discusses relevant works in the intersection among trust, recommendations systems, virtual communities, and agent-based systems. The target of the chapter is showing how, thanks to the use of trust-based solutions and artificial intelligent solutions like that understanding agents-based systems, the traditional recommender systems can improve the quality of their predictions. Moreover, when implemented as open multi-agent systems, trust-based recommender systems can efficiently support users of mobile virtual communities in searching for places, information, and items of interest.

Download Full-text

Trust-Aware Recommender Systems for Open and Mobile Virtual Communities

Collaborative Computer Security and Trust Management - Advances in Information Security, Privacy, and Ethics ◽

10.4018/978-1-60566-414-9.ch007 ◽

2010 ◽

pp. 135-153

Author(s):

N. Sahli ◽

G. Lenzini

Keyword(s):

Recommender Systems ◽

Virtual Communities ◽

Multi Agent Systems ◽

Artificial Intelligent ◽

Agent Based ◽

Agent Systems ◽

Multi Agent

This chapter surveys and discusses relevant works in the intersection among trust, recommendations systems, virtual communities, and agent-based systems. The target of the chapter is showing how, thanks to the use of trust-based solutions and artificial intelligent solutions like that understanding agents-based systems, the traditional recommender systems can improve the quality of their predictions. Moreover, when implemented as open multi-agent systems, trust-based recommender systems can efficiently support users of mobile virtual communities in searching for places, information, and items of interest.

Download Full-text

MULTI-AGENT LEARNING IN RECOMMENDER SYSTEMS FOR INFORMATION FILTERING ON THE INTERNET

International Journal of Cooperative Information Systems ◽

10.1142/s0218843001000266 ◽

2001 ◽

Vol 10 (01n02) ◽

pp. 81-100 ◽

Cited By ~ 9

Author(s):

JOAQUIN DELGADO ◽

NAOHIRO ISHII

Keyword(s):

Recommender Systems ◽

Information Filtering ◽

Target Function ◽

Majority Voting ◽

Information Goods ◽

Multi Agent Systems ◽

Agent Based ◽

Agent Learning ◽

Timely Fashion ◽

Multi Agent

Recommender Systems (RS), allow users to share information about items they like or dislike and obtain, in a timely fashion, recommendations based on predictions about unseen items (physical or information goods and/or services). In this process, users' preferences are considered to be the learning target functions. We study Agent-based Recommender Systems (ARS) under the scope of online learning in Multi-Agent systems (MAS). This approach models the problem as a pool of independent cooperative predictor agents, one per each user (the masters) in the system, in situations in which each agent (the learners) faces a sequence of trials, with a prediction to make in every step, eventually getting the correct value from its master. Each learner is willing to discover the degree of similarity among the target function of its master and those of other agents' masters (i.e. preference similarity). The agent uses this information for the calculation of its own prediction task, the goal being to make as few mistakes as possible. A simple, yet effective method is introduced in order to construct a compound algorithm for each agent by combining memory-based individual prediction and online weighted-majority voting. We give a theoretical mistake bound for this algorithm that is closely related to the total loss of the best predictor agent in the pool. Finally, we conduct some experiments obtaining results that empirically support these ideas and theories.

Download Full-text

Applications of multi-agent systems from the perspective of construction management: A literature review

Engineering Construction & Architectural Management ◽

10.1108/ecam-01-2021-0038 ◽

2021 ◽

Vol ahead-of-print (ahead-of-print) ◽

Author(s):

Liqun Xiang ◽

Yongtao Tan ◽

Geoffrey Shen ◽

Xin Jin

Keyword(s):

Literature Review ◽

Construction Management ◽

Future Research ◽

Multi Agent Systems ◽

Agent Based Models ◽

Content Type ◽

Agent Based ◽

Agent Systems ◽

Complex Problems ◽

Multi Agent

PurposeThe applications of multi-agent systems (MASs) are considered to be among the most promising paradigms for detailed investigations and reliable problem-solving methods, and MAS applications make it possible for researchers and practitioners to better understand complex systems. Although a number of prior studies have been conducted to address complex issues that arise from construction projects, few studies have summarised the applications and discussed the capacity of MASs from the perspective of construction management. To fill the gap, this paper provides a comprehensive literature review of MAS applications from the perspective of construction management.Design/methodology/approachWeb of Science and Scopus are the most commonly used international databases in conducting the literature reviews. A total of 86 relevant papers published in SCI-Expanded, SSCI and Ei Compendex journals related to the application of MASs from the perspective of construction management are selected to be analysed and discussed in this paper.FindingsBased on the 86 collected publications, the utilisations of MASs to support the management of the supply chain and the improvement of project performance are identified from the perspective of construction management, the characteristics and barriers of current MAS applications are analysed, a framework for developing agent-based models to address complex problems is proposed, and future research directions of MAS applications are discussed.Originality/valueThis review can serve as a useful reference for scholars to enhance their understanding of the current research and guide future research on MASs. The proposed framework can help build agent-based models to address complex problems in construction management.

Download Full-text

Distributed recommender system using multi-agent for social networks

10.32920/ryerson.14648478 ◽

2021 ◽

Author(s):

Lubaid Ahmed

Keyword(s):

Social Networks ◽

Social Network ◽

Recommender Systems ◽

Recommender System ◽

Network Simulator ◽

Clustering Methods ◽

Agent Based Simulation ◽

Agent Based ◽

Data Mining Algorithms ◽

Multi Agent

Social networks have become significant tools due to the vast and useful information existing in them. The social platforms also act as the storage of entered choices of millions of users for various applications such as political surveys, research studies, marketing product preferences and many more. Social network recommender systems exploit this information and direct users in selecting their choices. It is clear that recommender systems should be efficient enough to be able to process the huge magnitude of data that has been generated in recent years by social network users. This research proposes a foundation of an efficient and scalable recommender system to be able to process large amount of data (i.e. Big data) in a short amount of time. The main goal is providing scalability and efficiency of the recommender system. The simulation of the prototype of such a distributed recommender system by using multi-agent based technologies shows promising results. These prototypes provide recommendations to users about other users with the similar interests in online and distributed manner as real recommender systems. The agents can simulate users or can be used as the containers of algorithms for comparing the similarity between users by different approaches, such as cosine similarity and clustering methods for testing and examining real scenarios. To be able to test these prototypes in agent-based simulation environment an agent-based framework is developed. This framework has three modules named social network crawler, social network simulator and employed prototype of the distributed recommender system that use different text and data mining algorithms. Finally, newly developed performance metric (called Scalability Factor) is introduced that shows the minimum number of servers needed to be able to run the agent systems in parallel. This thesis shows using a distributed and parallel model for recommender systems is the key to increase the speed of recommendation convergence and as a result to provide scalability. Multi-agent based simulation results, coupled with numerical analysis affirm that the proposed solution provides scalability and efficiency for recommender systems.

Download Full-text

Distributed recommender system using multi-agent for social networks

10.32920/ryerson.14648478.v1 ◽

2021 ◽

Author(s):

Lubaid Ahmed

Keyword(s):

Social Networks ◽

Social Network ◽

Recommender Systems ◽

Recommender System ◽

Network Simulator ◽

Clustering Methods ◽

Agent Based Simulation ◽

Agent Based ◽

Data Mining Algorithms ◽

Multi Agent

Social networks have become significant tools due to the vast and useful information existing in them. The social platforms also act as the storage of entered choices of millions of users for various applications such as political surveys, research studies, marketing product preferences and many more. Social network recommender systems exploit this information and direct users in selecting their choices. It is clear that recommender systems should be efficient enough to be able to process the huge magnitude of data that has been generated in recent years by social network users. This research proposes a foundation of an efficient and scalable recommender system to be able to process large amount of data (i.e. Big data) in a short amount of time. The main goal is providing scalability and efficiency of the recommender system. The simulation of the prototype of such a distributed recommender system by using multi-agent based technologies shows promising results. These prototypes provide recommendations to users about other users with the similar interests in online and distributed manner as real recommender systems. The agents can simulate users or can be used as the containers of algorithms for comparing the similarity between users by different approaches, such as cosine similarity and clustering methods for testing and examining real scenarios. To be able to test these prototypes in agent-based simulation environment an agent-based framework is developed. This framework has three modules named social network crawler, social network simulator and employed prototype of the distributed recommender system that use different text and data mining algorithms. Finally, newly developed performance metric (called Scalability Factor) is introduced that shows the minimum number of servers needed to be able to run the agent systems in parallel. This thesis shows using a distributed and parallel model for recommender systems is the key to increase the speed of recommendation convergence and as a result to provide scalability. Multi-agent based simulation results, coupled with numerical analysis affirm that the proposed solution provides scalability and efficiency for recommender systems.

Download Full-text