Recommender systems are being improved by every means to be more accurate,more robust,and faster. Collaborative filtering is the mainstream type of recommendation algorithms,and its core is calculating the similarity ...
详细信息
Recommender systems are being improved by every means to be more accurate,more robust,and faster. Collaborative filtering is the mainstream type of recommendation algorithms,and its core is calculating the similarity between users or items based on *** recently found that the binary similarity based solely on who-rated-what rather than actual ratings output more accurate ***,from robust perspective,evaluated the binary collaborative filtering under multiple types of profile injection attacks on large dataset. Experimental results show binary collaborative filtering is more robust than actual ratings based collaborative filtering in all situations.
Aiming to overcome the shortcomings of P2P Grid safety,this paper presents and constructs a new P2P Grid trust model based on price motivate mechanism,and then designs and achieves an algorithm based on the *** penalt...
详细信息
Aiming to overcome the shortcomings of P2P Grid safety,this paper presents and constructs a new P2P Grid trust model based on price motivate mechanism,and then designs and achieves an algorithm based on the *** penalty factor of the algorithm can prevent the bad behavior and attack of malicious peers from systems;the time attenuation function reflects the trust value real and current situation of peers;the path attenuation function ensures the effectiveness and reliability of trust link;the price motivate mechanism ensure not only profit maximization for resource suppliers,but also cost minimization for resource *** simulation test verified that the algorithm reflecta the dynamics of P2P Grid trust,motivates the activity of the good peers and ensures thar the price motivate mechanism play a key regulation role in transaction market.
As one dimensional range profile is sensitive to direction,translation and intensity,it needs extra storage and big computation in target *** reduce storage and computation,a new multi-scale energy feature based on dy...
详细信息
As one dimensional range profile is sensitive to direction,translation and intensity,it needs extra storage and big computation in target *** reduce storage and computation,a new multi-scale energy feature based on dyadic wavelet transform is *** energy is not dependent on the translation of range profiles,the new feature has a translation invariable *** energy of two neighbor scales is divided to eliminate the intensity *** feature is also insensitive to direction *** the proposed feature was applied into real data for target identification,good result has been obtained,which is better than conventional methods. Therefore,the new method is an effective feature selection method for range profile identification.
This paper studies an actual and new setting of cost-sensitive learning,i.e.,combining test data with medical history under multiple-scale cost *** a new cost structure,an attribute selection strategy is incorporated ...
详细信息
This paper studies an actual and new setting of cost-sensitive learning,i.e.,combining test data with medical history under multiple-scale cost *** a new cost structure,an attribute selection strategy is incorporated to a lazy decision tree induction,so as to minimize the total cost on focused scale when medical history is dynamically utilized to current test *** experiments on six medical datasets in the UCI library demonstrate that the proposed lazy cost-sensitive decision tree algorithm has outperformed a group of existing costsensitive learning algorithms in a cost/budget-changing environment.
Nowadays,many organizations face the problem of how to accessing the legacy data by means of flexible mechanism. As a powerful and efficient solution to this problem,ontologybased data access(OBDA) technology enables ...
详细信息
Nowadays,many organizations face the problem of how to accessing the legacy data by means of flexible mechanism. As a powerful and efficient solution to this problem,ontologybased data access(OBDA) technology enables people to utilize data from heterogeneous sources through a well-defined ***,under the dynamic environment,the ontology which acts as concept view in an OBDA application may be changed and modified to meet users' ***,the changes occurred during ontology evolution process may be propagated across the whole *** order to minimize the influence of ontology evolution in OBDA system,we present a solution to control the impact of change propagation from ontology evolution,namely ICOE *** experiment illustrate that this method is helpful in managing an OBDA system.
Collaborative Filtering(CF) is one of the most successful recommender *** most critical step in CF is similarity *** CF,similarity is used for neighbor *** addition,it will be used as a weighted coefficient during the...
详细信息
Collaborative Filtering(CF) is one of the most successful recommender *** most critical step in CF is similarity *** CF,similarity is used for neighbor *** addition,it will be used as a weighted coefficient during the prediction ***,three different similarity measures are used:cosine based similarity,Pearson correlation coefficient based similarity and adjusted cosine based similarity. However,as these methods are based on linear correlations,they are also *** the linear correlation only takes into account the linear part of the *** paper introduces a new similarity measure based on mutual information to avoid the above *** experiments,done on MovieLens data sets,show that this new method outperforms traditional similarity measures under the nonlinear correlation circumstance.
This article studied on the rotation motion of trimaran unmanned surface *** experiment has been based on the trimaran unmanned surface vessel *** selected the Grenetic algorithm as the optimization method first, then...
详细信息
This article studied on the rotation motion of trimaran unmanned surface *** experiment has been based on the trimaran unmanned surface vessel *** selected the Grenetic algorithm as the optimization method first, then identified the test data of trimaran unmanned surface vessel rotation motion by the program assisted with ISIGHTFD *** compared the identified results with the test data and analyzed the error,we got the identified result matched the test result well from a set of compared data plots. The trimaran unmanned surface vessel model and its control system have important value for research and have good reference value for practical applications of unmanned surface
In order to increase transmission and processing speed,it is essential to compress the non-stationary leakage current in the insulator monitored *** Sensing(CS) combines sampling and compression with a small amount of...
详细信息
In order to increase transmission and processing speed,it is essential to compress the non-stationary leakage current in the insulator monitored *** Sensing(CS) combines sampling and compression with a small amount of sample to reconstruct signal well,which not only reduces hardware requirements,but also improves compression ***,computational complexity is high in CS recovery process of non-stationary *** this paper,CS combines with empirical mode decomposition method(EMD).Decompose Non-stationary leakage current into a finite number of stationary intrinsic mode functions (IMF),and then CS processes stationary *** experiment results indicate that it enhances not only the speed of process and operational efficiency but also compression ratio and reconstruction accuracy in the case that CS processes leakage current signal by EMD decomposition.
Modeling technology is the key to the advancement of road & bridge construction problem solving *** light of the existing defects of problem-solving modeling technology,a new function graphic modeling technology w...
详细信息
Modeling technology is the key to the advancement of road & bridge construction problem solving *** light of the existing defects of problem-solving modeling technology,a new function graphic modeling technology which can reflect problem solving characteristics is *** on function graphic object which has engineering properties of lengthvariable and unstructured,and can exist independently and work cooperatively with each other,a group of function graphic objects are used to emulate planed road & bridge construction sub-problem,and solving process is carried out interactively and dynamically by creating,operating,and deducing function graphic *** the proposed function graphic modeling technology,the traditional modeling method and realization model of road & bridge construction problem solving are broken through,and a new modeling and realization technology system which is suitable for solving characteristics of road & bridge construction problem is set up.
The capture of temporal dynamics of news streams has drawn increasing attentions in recent sequential data mining *** of them are based on the intuition that a "burst" of a topic is signaled by a growth of r...
详细信息
The capture of temporal dynamics of news streams has drawn increasing attentions in recent sequential data mining *** of them are based on the intuition that a "burst" of a topic is signaled by a growth of relevant words in a high intensity during a period of *** "burst features" can be efficiently identified by Kleinberg's two-state automaton model. The resolution is an important parameter of the *** affects the reliability of the results *** paper maximizes the reliability of the results by estimating adaptive resolution for each word with EM *** with the public news corpora prove that the unified resolution is a bottleneck of the performance,and the results with word-adaptive resolutions approximate to the maximum reliability well.
暂无评论