Limits...
A Dynamic Recommender System for Improved Web Usage Mining and CRM Using Swarm Intelligence.

Alphy A, Prabakaran S - ScientificWorldJournal (2015)

Bottom Line: Our dynamic recommender system was compared against traditional collaborative filtering systems.The results show that the proposed system has higher precision, coverage, F1 measure, and scalability than the traditional collaborative filtering systems.Moreover, the recommendations given by our system overcome the overspecialization problem by including variety in recommendations.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science and Engineering, SRM University, Chennai 603203, India.

ABSTRACT
In modern days, to enrich e-business, the websites are personalized for each user by understanding their interests and behavior. The main challenges of online usage data are information overload and their dynamic nature. In this paper, to address these issues, a WebBluegillRecom-annealing dynamic recommender system that uses web usage mining techniques in tandem with software agents developed for providing dynamic recommendations to users that can be used for customizing a website is proposed. The proposed WebBluegillRecom-annealing dynamic recommender uses swarm intelligence from the foraging behavior of a bluegill fish. It overcomes the information overload by handling dynamic behaviors of users. Our dynamic recommender system was compared against traditional collaborative filtering systems. The results show that the proposed system has higher precision, coverage, F1 measure, and scalability than the traditional collaborative filtering systems. Moreover, the recommendations given by our system overcome the overspecialization problem by including variety in recommendations.

No MeSH data available.


Clusters of agents obtained after applying Bluegill-BestPredictions algorithm.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4502334&req=5

fig5: Clusters of agents obtained after applying Bluegill-BestPredictions algorithm.

Mentions: Figure 5 shows the clusters of agents obtained after applying Bluegill-BestPredictions algorithm. In the Bluegill-BestPredictions algorithm for higher density cluster the Maxth value ranges from 0.75 to 1.0. For lower density cluster we set the Minth value as 0.40. In medium density cluster the similarity lies between 0.40 and 0.75. In high density cluster high_range means similarity above 75% of Maxth. In lower density cluster, high_range means similarity above 75% of Minth, mid_range means similarity between 45% and 75% of Minth, and low_range means similarity below 45% of Minth. In medium density cluster high_range means similarity above 75% of obtained medium density cluster similarity value. Here, mid_range lies within 45% and 75% of medium density cluster similarity value. These parameters are set by trial and error method. In Figure 5 Clus_threshold = 10 and Url_count  _threshold = ICTF∗Clus_threshold [19], where ICTF denotes item count threshold frequency [19] that represents the minimum number of URL patterns that represent that session. Here ICTF = 0.10. An ICTF value is a real number and it lies between 0 and 1. Clus_threshold represents the minimum cluster size required for a valid cluster.


A Dynamic Recommender System for Improved Web Usage Mining and CRM Using Swarm Intelligence.

Alphy A, Prabakaran S - ScientificWorldJournal (2015)

Clusters of agents obtained after applying Bluegill-BestPredictions algorithm.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4502334&req=5

fig5: Clusters of agents obtained after applying Bluegill-BestPredictions algorithm.
Mentions: Figure 5 shows the clusters of agents obtained after applying Bluegill-BestPredictions algorithm. In the Bluegill-BestPredictions algorithm for higher density cluster the Maxth value ranges from 0.75 to 1.0. For lower density cluster we set the Minth value as 0.40. In medium density cluster the similarity lies between 0.40 and 0.75. In high density cluster high_range means similarity above 75% of Maxth. In lower density cluster, high_range means similarity above 75% of Minth, mid_range means similarity between 45% and 75% of Minth, and low_range means similarity below 45% of Minth. In medium density cluster high_range means similarity above 75% of obtained medium density cluster similarity value. Here, mid_range lies within 45% and 75% of medium density cluster similarity value. These parameters are set by trial and error method. In Figure 5 Clus_threshold = 10 and Url_count  _threshold = ICTF∗Clus_threshold [19], where ICTF denotes item count threshold frequency [19] that represents the minimum number of URL patterns that represent that session. Here ICTF = 0.10. An ICTF value is a real number and it lies between 0 and 1. Clus_threshold represents the minimum cluster size required for a valid cluster.

Bottom Line: Our dynamic recommender system was compared against traditional collaborative filtering systems.The results show that the proposed system has higher precision, coverage, F1 measure, and scalability than the traditional collaborative filtering systems.Moreover, the recommendations given by our system overcome the overspecialization problem by including variety in recommendations.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science and Engineering, SRM University, Chennai 603203, India.

ABSTRACT
In modern days, to enrich e-business, the websites are personalized for each user by understanding their interests and behavior. The main challenges of online usage data are information overload and their dynamic nature. In this paper, to address these issues, a WebBluegillRecom-annealing dynamic recommender system that uses web usage mining techniques in tandem with software agents developed for providing dynamic recommendations to users that can be used for customizing a website is proposed. The proposed WebBluegillRecom-annealing dynamic recommender uses swarm intelligence from the foraging behavior of a bluegill fish. It overcomes the information overload by handling dynamic behaviors of users. Our dynamic recommender system was compared against traditional collaborative filtering systems. The results show that the proposed system has higher precision, coverage, F1 measure, and scalability than the traditional collaborative filtering systems. Moreover, the recommendations given by our system overcome the overspecialization problem by including variety in recommendations.

No MeSH data available.