A two-step personalized location recommendation based on multi-objective immune algorithm

Geng, Bingrui; Jiao, Licheng; Gong, Maoguo; Wu, Yue

doi:10.1016/j.ins.2018.09.068

Cited by 32 publications

(13 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are many parameters in the algorithm. Refer to the parameter settings in Literature [15]. The values of parameters are shown in Table 1. (1) Experiments for the nearest neighbor K…”

Section: Analysis Of Experimental Resultsmentioning

confidence: 99%

“…In the minimization optimization problems, if and only if  i  {1,2,…, }, i (x)  f i (y), and  j{1,2,…, }, j (x) < f j (y), which is called x dominates y. If a solution is called the Pareto-optimal solution, it should not be dominated by other solutions [15]. For a multi-objective optimization problem, if the solution in a set is non-dominated with each other, the set is called Pareto-optimal solution set.…”

Section: Multi-objective Optimization Algorithmmentioning

confidence: 99%

“…The accuracy is the proportion of the items that the user likes in the recommendation list, the calculation is formula (14). The recall rate is how many items the user likes in the test set appear in the recommendation list, the calculation is formula (15). As one of the commonly used evaluation Indicators in the recommendation algorithm, MAE measures the accuracy of the prediction by calculating the average deviation between the user's predicted ratings and the actual ratings.…”

Section: Evaluation Indicatorsmentioning

confidence: 99%

See 2 more Smart Citations

Multi-Objective Optimization Recommendation Algorithm Based on Collaborative Filtering and Item Similarity

2019

Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

The traditional collaborative filtering recommendation algorithm generates recommendations for users by calculating the similarity between users or items, without considering other factors and data sparsity problem. The appearance of personalized tags can reflect the characteristics of items. Some algorithms try to optimize the recommendation results by setting weight coefficients of various factors. However, it is unreasonable to set the same weight coefficients for different users. To solve these problems, we propose a multi-objective optimization recommendation algorithm based on collaborative filtering and item similarity. First. The algorithm calculates the similarity between users by improved user-based collaborative filtering recommendation algorithm and generates a candidate recommendation set. Then, the algorithm uses the coupled object similarity measure method to calculate the similarity between items. Finally, we use NNIA algorithm to optimize the former two goals and generate the final recommendation results. Experimental results on the Movielens dataset shows that the proposed algorithm has good recommendation effect.

show abstract

“…There are many parameters in the algorithm. Refer to the parameter settings in Literature [15]. The values of parameters are shown in Table 1. (1) Experiments for the nearest neighbor K…”

Section: Analysis Of Experimental Resultsmentioning

confidence: 99%

Section: Multi-objective Optimization Algorithmmentioning

confidence: 99%

Section: Evaluation Indicatorsmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Objective Optimization Recommendation Algorithm Based on Collaborative Filtering and Item Similarity

2019

Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

show abstract

“…AISs with hypermutation operators have achieved great success in solving complex optimization problems, including single-objective and multi-objective optimization, for example, (Shang et al 2011;Yao et al 2016;Alizadeh, Meskin, and Khorasani 2017;Dudek 2017;Lin et al 2018;Xu et al 2018;Geng et al 2019). Compared to successes in application, the theoretical analysis on understanding the optimization behavior of hypermutation operators is underdeveloped, particularly on optimizing multi-objective optimization problems (MOPs).…”

Section: Introductionmentioning

confidence: 99%

Runtime Analysis of Somatic Contiguous Hypermutation Operators in MOEA/D Framework

Huang

Zhou

2020

AAAI

View full text Add to dashboard Cite

Somatic contiguous hypermutation (CHM) operators are important variation operators in artificial immune systems. The few existing theoretical studies are only concerned with understanding the optimization behavior of CHM operators on solving single-objective optimization problems. The MOEA/D framework is one of the most popular strategies for solving multi-objective optimization problems (MOPs). In this paper, we present a runtime analysis of using two CHM operators in MOEA/D framework for solving five benchmark MOPs, including four bi-objective and one many-objective problems. Our analyses show that the expected runtimes of CHM operators on the four bi-objective problems are better than or as good as that of the well-studied standard bit mutation operator. Moreover, using CHM operators in MOEA/D framework can improve the best known upper bound on the many-objective problem by a factor of n. This paper provides insight into understanding the optimization behavior of CHM operators in the well-known MOEA/D framework, and indicates that using the CHM operator in MOEA/D framework is a promising method for handling MOPs.

show abstract

“…Recent studies have applied collaborative filtering, content‐based methods, and hybrid to extract useful information from multiple types of data for modeling user preferences. Simultaneously combining multiple information by some means may help alleviate the problems caused by sparse data, which however has proven to be a significant challenge (Geng, Jiao, Gong, Li, & Wu, ).…”

Section: Introductionmentioning

confidence: 99%

Location recommendation by combining geographical, categorical, and social preferences with location popularity

Mao

2019

Proceedings of the Association for Information Science and Tech

View full text Add to dashboard Cite

Location recommendation could not only increase location visibility, but free users from information overload dilemma. We propose a novel location recommendation method, which first exploits geographical proximity of locations and check‐in history of users' friends to measure geographical preference and social preference respectively, and innovatively estimates categorical preference by calculating semantic similarity between location tags based on category hierarchy. Then, all of three preferences are integrated, and location popularity is introduced as a coefficient to determine the priority of locations to be recommended. The preliminary result on a publicly available real‐word dataset shows that the proposed method significantly outperforms two state‐of‐the‐art baselines. This study expands the methods of location correlation measurement, and can meet user's personalized information needs.

show abstract

A two-step personalized location recommendation based on multi-objective immune algorithm

Cited by 32 publications

References 27 publications

Multi-Objective Optimization Recommendation Algorithm Based on Collaborative Filtering and Item Similarity

Multi-Objective Optimization Recommendation Algorithm Based on Collaborative Filtering and Item Similarity

Runtime Analysis of Somatic Contiguous Hypermutation Operators in MOEA/D Framework

Location recommendation by combining geographical, categorical, and social preferences with location popularity

Contact Info

Product

Resources

About