Handling Natural Noise in Multi Criteria Recommender System utilizing effective similarity measure and Particle Swarm Optimization

Choudhary, Priyankar; Kant, Vibhor; Dwivedi, Pragya

doi:10.1016/j.procs.2017.09.168

“…The tendency measures the level of diversity that users prefers. Therefore, the objective of RS algorithm is calculated using Equation (15). 39 Terms in the equations have been defined in earlier sections.…”

Section: Objective Functionsmentioning

confidence: 99%

“…Moreover, the proposed method is also evaluated by how well they balance between the multiple RS quality factors. The diversity, novelty, and user tendency evaluation criteria are measured by Equations ( 11), ( 12), (15), respectively. For optimization-based CF (NSGAII-CF, MOPSO-CF, MOBFO-CF, MCLBFO-CF, HSMBFO-CF), the obtained Pareto frontier represents a set of optimal solutions.…”

Section: Comparison On Movie Recommendation Taskmentioning

confidence: 99%

“…The recommendation performance criteria are regarded as optimization goals. For example, [15][16][17][18][19] use particle swarm optimization (PSO) and evolutionary algorithm (EA) to balance the technique performance in accuracy and diversity, or accuracy and item unpopularity. However, these heuristic optimization algorithms are probably confined in the local optimum thus not able find global optimal solution.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multicriteria recommendation based on bacterial foraging optimization

Geng

¹

,

He

²

,

Wang

³

et al. 2021

Int J of Intelligent Sys

View full text Add to dashboard Cite

Recommender systems assist users to make decisions among a huge volume of options. Accuracy-oriented recommender systems focus on the prediction power of algorithms and neglect that users may appreciate diverse and novel recommendations in real-world scenarios. Thus, this paper proposed a multicriteria recommendation model that can optimize the recommendation accuracy, diversity, novelty, and individual tendency simultaneously.Additionally, a new multiobjective bacterial foraging optimization method is proposed to improve its searching capability and the performance of recommendation model. The proposed optimization-based multicriteria recommendation algorithm is compared with existing methods on both benchmark functions and real-world data sets. The results demonstrate that the proposed algorithm is superior to other recommendation algorithms in most cases. This study provides insights in recommendation system design and draws scholarly attention to the optimization-based recommendation strategy.

show abstract

“…Moreover, in addition to using fuzzy approach, research has recently begun to apply some optimisation algorithms such as PSO algorithms for modelling the criteria ratings have started evolving. In Choudhary et al (2017a), PSO algorithm together with some effective similarity measures has been used for handling natural noise in multi-criteria RSs and for improving their accuracy. Furthermore, PSO algorithm was used in Choudhary et al (2017b) for learning optimal weights of the similarities between the criteria ratings to estimate the overall rating.…”

Section: Related Workmentioning

confidence: 99%

Evaluating the performance of a neural network-based multi-criteria recommender system

Hassan

¹

,

Hamada

²

2019

IJSTDS

View full text Add to dashboard Cite

Frequent use of internet applications and rapid increase in volumes of resources have made it difficult for online users to effectively make decisions on the kinds of information or items to select. Recommender systems (RSs) are intelligent decision-support tools that exploit preferences of users and suggest items that might be interesting to them. RSs are one of the various solutions proposed to address the problems of information overload. Traditionally, RSs use single rating techniques to predict and represent preferences of users for items that are not yet seen. Multi-criteria RSs use multiple ratings to various attributes of items for improving prediction and recommendation accuracy of the systems. However, one major challenge of multi-criteria RSs is the choice of an efficient approach for modelling the criteria ratings. Therefore, this paper aimed at employing artificial neural networks to model the criteria ratings and determine the predictive performance of the systems based on aggregation function approach. Seven evaluation metrics have been used to evaluate and measure the accuracy of the systems. The empirical results of the study have shown that the proposed technique has the highest prediction and recommendation than the corresponding traditional technique.Reference to this paper should be made as follows: Hassan, M. and Hamada, M. (2019) 'Evaluating the performance of a neural network-based multi-criteria recommender system', Int.Evaluating the performance of a neural network-based multi-criteria 55 Mohamed Hamada is currently a Senior Associate Professor working in the Software Engineering Lab, at the University of Aizu, Aizuwakamatsu, Fukushima, Japan. He obtained his PhD from the University of Tsukuba, Japan. His research interest is in artificial intelligence and the study of functional logic languages, natural inspired computation, cloud computing-based e-learning and m-learning systems, multimedia, and web-based and advanced learning technologies. He is also interested in smart devices (such as smartphones and tablets) applications development and innovation, and so on. He is a regular Visiting Professor at Fatih University, Istanbul, Turkey and the African This paper is a revised and expanded version of a paper entitled 'Performance analysis of neural networks-based multi-criteria recommender systems' presented at

show abstract

“…Moreover, in addition to using fuzzy approach, research has recently begun to apply some optimisation algorithms such as PSO algorithms for modelling the criteria ratings have started evolving. In Choudhary et al (2017a), PSO algorithm together with some effective similarity measures has been used for handling natural noise in multi-criteria RSs and for improving their accuracy. Furthermore, PSO algorithm was used in Choudhary et al (2017b) for learning optimal weights of the similarities between the criteria ratings to estimate the overall rating.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

Evaluating the performance of a neural network-based multi-criteria recommender system

Hassan¹,

Hamada²

2019

IJSTDS

View full text Add to dashboard Cite

Frequent use of internet applications and rapid increase in volumes of resources have made it difficult for online users to effectively make decisions on the kinds of information or items to select. Recommender systems (RSs) are intelligent decision-support tools that exploit preferences of users and suggest items that might be interesting to them. RSs are one of the various solutions proposed to address the problems of information overload. Traditionally, RSs use single rating techniques to predict and represent preferences of users for items that are not yet seen. Multi-criteria RSs use multiple ratings to various attributes of items for improving prediction and recommendation accuracy of the systems. However, one major challenge of multi-criteria RSs is the choice of an efficient approach for modelling the criteria ratings. Therefore, this paper aimed at employing artificial neural networks to model the criteria ratings and determine the predictive performance of the systems based on aggregation function approach. Seven evaluation metrics have been used to evaluate and measure the accuracy of the systems. The empirical results of the study have shown that the proposed technique has the highest prediction and recommendation than the corresponding traditional technique.

show abstract

Handling Natural Noise in Multi Criteria Recommender System utilizing effective similarity measure and Particle Swarm Optimization

Cited by 15 publications

References 14 publications

Multicriteria recommendation based on bacterial foraging optimization

Multicriteria recommendation based on bacterial foraging optimization

Evaluating the performance of a neural network-based multi-criteria recommender system

Evaluating the performance of a neural network-based multi-criteria recommender system

Contact Info

Product

Resources

About