In most current recommender systems,the goal to accurately predict what people want leads to the tendency to recommend popular items,which is less helpful in revealing user＇s personality,especially to new users.In th...In most current recommender systems,the goal to accurately predict what people want leads to the tendency to recommend popular items,which is less helpful in revealing user＇s personality,especially to new users.In this paper,we propose a heuristic music recommendation method for niche market by focusing on how to identify user＇s personality as soon as possible.Instead of trying to improve algorithm＇s performance on new users by recommending the most popular items,we work on how to make them ＂familiar＂ with the system earlier.The method is more suitable for brand-new users,and gives a hint to solve the cold start problem.In real applications it is better to combine it with a traditional approach.展开更多
Personalization is the adaptation of the services to fit the user’s interests,characteristics and needs.The key to effective personalization is user profiling.Apart from traditional collaborative and content-based ap...Personalization is the adaptation of the services to fit the user’s interests,characteristics and needs.The key to effective personalization is user profiling.Apart from traditional collaborative and content-based approaches,a number of classification and clustering algorithms have been used to classify user related information to create user profiles.However,they are not able to achieve accurate user profiles.In this paper,we present a new clustering algorithm,namely Multi-Dimensional Clustering(MDC),to determine user profiling.The MDC is a version of the Instance-Based Learner(IBL)algorithm that assigns weights to feature values and considers these weights for the clustering.Three feature weight methods are proposed for the MDC and,all three,have been tested and evaluated.Simulations were conducted with using two sets of user profile datasets,which are the training(includes 10,000 instances)and test(includes 1000 instances)datasets.These datasets reflect each user’s personal information,preferences and interests.Additional simulations and comparisons with existing weighted and non-weighted instance-based algorithms were carried out in order to demonstrate the performance of proposed algorithm.Experimental results using the user profile datasets demonstrate that the proposed algorithm has better clustering accuracy performance compared to other algorithms.This work is based on the doctoral thesis of the corresponding author.展开更多
Recent progress of Web 2.0 applications has witnessed the rapid development of microblog in China, which has already been one of the most important ways for online communications, especially on sharing information. Th...Recent progress of Web 2.0 applications has witnessed the rapid development of microblog in China, which has already been one of the most important ways for online communications, especially on sharing information. This paper tries to make an in-depth investigation on the big data modeling and analysis of microblog ecosystem in China by using a real dataset containing over17 million records of SinaWeibo users. First, we present the detailed geography, gender, authentication, education and age analysis of microblog users in this dataset. Then we conduct the numerical features distribution analysis, propose the user influence formula and calculate the influences for different kinds of microblog users. Finally, user content intention analysis is performed to reveal users most concerns in their daily life.展开更多
基金supported by the National Natural Science Foundation of China under Grant Nos.60973120,60903073,61003231,61103109,and11105024
文摘In most current recommender systems,the goal to accurately predict what people want leads to the tendency to recommend popular items,which is less helpful in revealing user＇s personality,especially to new users.In this paper,we propose a heuristic music recommendation method for niche market by focusing on how to identify user＇s personality as soon as possible.Instead of trying to improve algorithm＇s performance on new users by recommending the most popular items,we work on how to make them ＂familiar＂ with the system earlier.The method is more suitable for brand-new users,and gives a hint to solve the cold start problem.In real applications it is better to combine it with a traditional approach.
文摘Personalization is the adaptation of the services to fit the user’s interests,characteristics and needs.The key to effective personalization is user profiling.Apart from traditional collaborative and content-based approaches,a number of classification and clustering algorithms have been used to classify user related information to create user profiles.However,they are not able to achieve accurate user profiles.In this paper,we present a new clustering algorithm,namely Multi-Dimensional Clustering(MDC),to determine user profiling.The MDC is a version of the Instance-Based Learner(IBL)algorithm that assigns weights to feature values and considers these weights for the clustering.Three feature weight methods are proposed for the MDC and,all three,have been tested and evaluated.Simulations were conducted with using two sets of user profile datasets,which are the training(includes 10,000 instances)and test(includes 1000 instances)datasets.These datasets reflect each user’s personal information,preferences and interests.Additional simulations and comparisons with existing weighted and non-weighted instance-based algorithms were carried out in order to demonstrate the performance of proposed algorithm.Experimental results using the user profile datasets demonstrate that the proposed algorithm has better clustering accuracy performance compared to other algorithms.This work is based on the doctoral thesis of the corresponding author.
基金supported by National Natural Science Foundation of China(No.61272362)National Basic Research Program ofChina(973 Program)(No.2013CB329606)High-Tech Development Plan of Xinjiang(No.201212124)
文摘Recent progress of Web 2.0 applications has witnessed the rapid development of microblog in China, which has already been one of the most important ways for online communications, especially on sharing information. This paper tries to make an in-depth investigation on the big data modeling and analysis of microblog ecosystem in China by using a real dataset containing over17 million records of SinaWeibo users. First, we present the detailed geography, gender, authentication, education and age analysis of microblog users in this dataset. Then we conduct the numerical features distribution analysis, propose the user influence formula and calculate the influences for different kinds of microblog users. Finally, user content intention analysis is performed to reveal users most concerns in their daily life.
