共查询到20条相似文献，搜索用时 609 毫秒
1.
《Journal of The Franklin Institute》2022,359(3):1362-1381
Error entropy is a well-known learning criterion in information theoretic learning (ITL), and it has been successfully applied in robust signal processing and machine learning. To date, many robust learning algorithms have been devised based on the minimum error entropy (MEE) criterion, and the Gaussian kernel function is always utilized as the default kernel function in these algorithms, which is not always the best option. To further improve learning performance, two concepts using a mixture of two Gaussian functions as kernel functions, called mixture error entropy and mixture quantized error entropy, are proposed in this paper. We further propose two new recursive least-squares algorithms based on mixture minimum error entropy (MMEE) and mixture quantized minimum error entropy (MQMEE) optimization criteria. The convergence analysis, steady-state mean-square performance, and computational complexity of the two proposed algorithms are investigated. In addition, the reason why the mixture mechanism (mixture correntropy and mixture error entropy) can improve the performance of adaptive filtering algorithms is explained. Simulation results show that the proposed new recursive least-squares algorithms outperform other RLS-type algorithms, and the practicality of the proposed algorithms is verified by the electro-encephalography application. 相似文献
2.
Some of the most popular measures to evaluate information filtering systems are usually independent of the users because they are based in relevance judgments obtained from experts. On the other hand, the user-centred evaluation allows showing the different impressions that the users have perceived about the system running. This work is focused on discussing the problem of user-centred versus system-centred evaluation of a Web content personalization system where the personalization is based on a user model that stores long term (section, categories and keywords) and short term interests (adapted from user provided feedback). The user-centred evaluation is based on questionnaires filled in by the users before and after using the system and the system-centred evaluation is based on the comparison between ranking of documents, obtained from the application of a multi-tier selection process, and binary relevance judgments collected previously from real users. The user-centred and system-centred evaluations performed with 106 users during 14 working days have provided valuable data concerning the behaviour of the users with respect to issues such as document relevance or the relative importance attributed to different ways of personalization. The results obtained shows general satisfaction on both the personalization processes (selection, adaptation and presentation) and the system as a whole. 相似文献
3.
《Journal of The Franklin Institute》2023,360(9):6515-6536
The performance of the current state estimation will degrade in the existence of slow-varying noise statistics. To solve the aforementioned issues, an improved strong tracking maximum correntropy criterion variational-Bayesian adaptive Kalman filter is presented in this paper. First of all, the inverse-Wishart distribution, as the conjugate-prior, is adopted to model the unknown and time-varying measurement and process noise covariances, then the noise covariances and system state are estimated via the variational Bayesian method. Secondly, the multiple fading-factors are obtained and evaluated to modify the prediction error covariance matrix to address the problems associated with inaccurate error estimation. Finally, the maximum correntropy criterion is employed to correct the filtering gain, which improves the filtering performance of the proposed algorithm. Simulation results show that the proposed filter exhibits better accuracy and convergence performance compared to other existing algorithms. 相似文献
4.
This paper proposes collaborative filtering as a means to predict semantic preferences by combining information on social ties with information on links between actors and semantics. First, the authors present an overview of the most relevant collaborative filtering approaches, showing how they work and how they differ. They then compare three different collaborative filtering algorithms using articles published by New York Times journalists from 2003 to 2005 to predict preferences, where preferences refer to journalists’ inclination to use certain words in their writing. Results show that while preference profile similarities in an actor’s neighbourhood are a good predictor of her semantic preferences, information on her social network adds little to prediction accuracy. 相似文献
5.
Guobin Chang Tianhe Xu Chao Chen Bing Ji Shengquan Li 《Journal of The Franklin Institute》2019,356(9):4928-4947
Carrier-smoothing-code filtering (CSCF) is widely used in GNSS signal processing to combine code pseudoranges and carrier phases. Position-domain (PD) CSCF is generally more accurate and less sensitive to visible satellite changes than range-domain (RD) CSCF. However, PD-CSCF necessitates at least four visible satellites. Intermittent satellite deficiency with less than four visible satellites is not uncommon in harsh environments like urban canyons. At such deficiency epochs, the PD-CSCF convergence has to break off. This study aims to bridge intermittent deficiency epochs in PD-CSCF without introducing any external information. The proposed solution is called switching RD and PD CSCF in which PD filter is replaced by RD filter at deficiency epochs. Besides detailing the seamless switching algorithms from PD/RD to RD/PD filters, a global RD filter, different from the conventional one, is developed to preserve the correlations among smoothed pseudoranges corresponding to different satellites. Compared to the conventional PD and RD CSCF algorithms, smoother results can be expected from the proposed switching filter, especially after the intermittent deficiency epochs. Experiments are conducted using real BDS signals. Cases with different kinds of deficiency are considered. Superiority of the proposed method is clearly observed from the results. 相似文献
6.
User-model based personalized summarization 总被引：3，自引：0，他引：3
The potential of summary personalization is high, because a summary that would be useless to decide the relevance of a document if summarized in a generic manner, may be useful if the right sentences are selected that match the user interest. In this paper we defend the use of a personalized summarization facility to maximize the density of relevance of selections sent by a personalized information system to a given user. The personalization is applied to the digital newspaper domain and it used a user-model that stores long and short term interests using four reference systems: sections, categories, keywords and feedback terms. On the other side, it is crucial to measure how much information is lost during the summarization process, and how this information loss may affect the ability of the user to judge the relevance of a given document. The results obtained in two personalization systems show that personalized summaries perform better than generic and generic-personalized summaries in terms of identifying documents that satisfy user preferences. We also considered a user-centred direct evaluation that showed a high level of user satisfaction with the summaries. 相似文献
7.
The input-output finite-time filtering problem is addressed for a class of switched linear parameter-varying systems in this paper. Firstly, by constructing a parameter-dependent Lyapunov function and resorting to the average dwell time approach, sufficient conditions ensuring finite-time boundedness and input-output finite-time stability are established for the augmented filtering error system. Then, a parameter-dependent asynchronous filter is designed such that the augmented filtering error system are both finite-time bounded and input-output finite-time stable. Finally, the active magnetic bearing model is introduced and verifies the main algorithms in this paper. 相似文献
8.
Shayan A. Tabrizi Azadeh Shakery Hamed Zamani Mohammad Ali Tavallaei 《Information processing & management》2018,54(4):630-656
Despite the importance of personalization in information retrieval, there is a big lack of standard datasets and methodologies for evaluating personalized information retrieval (PIR) systems, due to the costly process of producing such datasets. Subsequently, a group of evaluation frameworks (EFs) have been proposed that use surrogates of the PIR evaluation problem, instead of addressing it directly, to make PIR evaluation more feasible. We call this group of EFs, indirect evaluation frameworks. Indirect frameworks are designed to be more flexible than the classic (direct) ones and much cheaper to be employed. However, since there are many different settings and methods for PIR, e.g., social-network-based vs. profile-based PIR, and each needs some special kind of data to do the personalization based on, not all the evaluation frameworks are applicable to all the PIR methods. In this paper, we first review and categorize the frameworks that have already been introduced for evaluating PIR. We further propose a novel indirect EF based on citation networks (called PERSON), which allows repeatable, large-scale, and low-cost PIR experiments. It is also more information-rich compared to the existing EFs and can be employed in many different scenarios. The fundamental idea behind PERSON is that in each document (paper) d, the cited documents are generally related to d from the perspective of d’s author(s). To investigate the effectiveness of the proposed EF, we use a large collection of scientific papers. We conduct several sets of experiments and demonstrate that PERSON is a reliable and valid EF. In the experiments, we show that PERSON is consistent with the traditional Cranfield-based evaluation in comparing non-personalized IR methods. In addition, we show that PERSON can correctly capture the improvements made by personalization. We also demonstrate that its results are highly correlated with those of another salient EF. Our experiments on some issues about the validity of PERSON also show its validity. It is also shown that PERSON is robust w.r.t. its parameter settings. 相似文献
9.
This paper studies the distributed Kalman consensus filtering problem based on the event-triggered (ET) protocol for linear discrete time-varying systems with multiple sensors. The ET strategy of the send-on-delta rule is employed to adjust the communication rate during data transmission. Two series of Bernoulli random variables are introduced to represent the ET schedules between a sensor and an estimator, and between an estimator and its neighbor estimators. An optimal distributed filter with a given recursive structure in the linear unbiased minimum variance criterion is derived, where solution of cross-covariance matrix (CCM) between any two estimators increases the complexity of the algorithm. In order to avert CCM, a suboptimal ET Kalman consensus filter is also presented, where the filter gain and the consensus gain are solved by minimizing an upper bound of filtering error covariance. Boundedness of the proposed suboptimal filter is analyzed based on a Lyapunov function. A numerical simulation verifies the effectiveness of the proposed algorithms. 相似文献
10.
《Journal of The Franklin Institute》2022,359(3):1274-1302
This article proposes an affine-projection-like maximum correntropy (APLMC) algorithm for robust adaptive filtering. The proposed APLMC algorithm is derived by using the objective function based on the maximum correntropy criterion (MCC), which can availably suppress the bad effects of impulsive noise on filter weight updates. But the overall performance of the APLMC algorithm may be decreased when the input signal is polluted by noise. To compensate for the deviation of the APLMC algorithm in the input noise interference environment, the bias compensation (BC) method is introduced. Therefore, the bias-compensated APLMC (BC-APLMC) algorithm is presented. Besides, the convergence of the BC-APLMC algorithm in the mean and the mean square sense is studied, which provides a constraint range for the step-size. Computer simulation results show that the APLMC, and BC-APLMC algorithms are valid in acoustic echo cancellation and system identification applications. It also shows that the proposed algorithms are robust in the presence of input noise and impulse noise. 相似文献
11.
《Journal of The Franklin Institute》2022,359(8):3951-3978
This paper presents explicit and implicit discrete-time realizations for the robust exact filtering differentiator, aiming to facilitate an adequate posterior implementation structure in digital devices. This paper firstly presents an analysis of an explicit discrete-time realization of the filtering differentiator based on linear systems’ exact discretization with a zero-order holder. For this case, however, high-order terms in the filter dynamics may cause instability of the estimation error for signals with unbounded derivatives. Hence, two other new discrete-time realizations of the filtering differentiator are derived by removing some high-order terms in the filter dynamics. The first one is an explicit discrete-time realization, while the second one is implicit. After a finite time, both preserve the accuracy of the continuous-time robust exact filtering differentiator in the presence of measurement noise. For each proposed discrete-time scheme, a stability analysis based on homogeneity is provided. Finally, the simulation results include comparisons between the proposed implicit and explicit discrete-time realizations with other existing schemes. These numerical studies highlight that the implicit scheme supersedes the explicit one, consistent with the implicit and explicit realizations of other continuous-time algorithms. 相似文献
12.
《Information processing & management》1999,35(5):633-654
New methods and new systems are needed to filter or to selectively distribute the increasing volume of electronic information being produced nowadays. An effective information filtering system is one that provides the exact information that fulfills user's interests with the minimum effort by the user to describe it. Such a system will have to be adaptive to the user changing interest. In this paper we describe and evaluate a learning model for information filtering which is an adaptation of the generalized probabilistic model of Information Retrieval. The model is based on the concept of `uncertainty sampling', a technique that allows for relevance feedback both on relevant and nonrelevant documents. The proposed learning model is the core of a prototype information filtering system called ProFile. 相似文献
13.
The prior studies on information disclosure in location-based services (LBS) suggested that the perceived benefits of information disclosure in LBS were manifested by three benefits, namely, locatability, personalization, and social benefits. The three benefits might affect information disclosure intention differently. As an extension, individual factors, such as gender, may affect the relationship. However, according to literature, little research has investigated on the combined influence of the three benefits on the information disclosure intention in LBS with the gender as a moderator. Based upon the self-determination and social role theories, this study intends to bridge the gap empirically. The hypotheses are largely supported by 215 respondents. Unexpectedly, the research findings show that for females, locatability and personalization are more important in predicting their information disclosure intention, whereas for males, the social benefit has more of an impact on information disclosure intention, which is opposite to the hypotheses and convention. Furthermore, the research findings indicate that the behaviors of males and females may conform to the roles distributed within a society of this information age rather than to the personalities of the individuals. Finally, the implications are presented. 相似文献
14.
由于缺少SAR散射波干扰真实数据,建立了SAR散射波干扰的物理模型,分析了地面散射点的分布特征,给出了仿真结果.结果表明,对于极度粗糙地表和中等粗糙地表,散射波干扰充斥于SAR的整个波束,不能采用空间滤波的方式进行干扰抑制;对于中等粗糙地表,增加快时间采样点数,可以增加相关散射点的数量,因此可以用空间-快时间自适应滤波的方法提高干扰抑制性能. 相似文献
15.
In the face of ubiquitous information communication technology, the presence of blogs, personal websites, and public message
boards give the illusion of uncensored criticism and discussion of the ethical implications of business activities. However,
little attention has been paid to the limitations on free speech posed by the control of access to the Internet by private
entities, enabling them to censor content that is deemed critical of corporate or public policy. The premise of this research
is that transparency alone will not achieve the desired results if ICT is used in a one way system, controlled by the provider
of information. Stakeholders must have an avenue using the same technology to respond to and interact with the information.
We propose a model that imposes on corporations a public trust, requiring these gatekeepers of communication technology to
preserve individual rights to criticism and review. 相似文献
16.
Feiyan Chen Feng Ding Ling Xu Tasawar Hayat 《Journal of The Franklin Institute》2018,355(7):3381-3398
For multivariable systems with autoregressive moving average noises, we decompose the multivariable system into m subsystems (m denotes the number of outputs) and present a maximum likelihood generalized extended gradient algorithm and a data filtering based maximum likelihood extended gradient algorithm to estimate the parameter vectors of these subsystems. By combining the maximum likelihood principle and the data filtering technique, the proposed algorithms are effective and have computational advantages over existing estimation algorithms. Finally, a numerical simulation example is given to support the developed methods and to show their effectiveness. 相似文献
17.
It has been argued that the Internet and social media increase the number of available viewpoints, perspectives, ideas and opinions available, leading to a very diverse pool of information. However, critics have argued that algorithms used by search engines, social networking platforms and other large online intermediaries actually decrease information diversity by forming so-called “filter bubbles”. This may form a serious threat to our democracies. In response to this threat others have developed algorithms and digital tools to combat filter bubbles. This paper first provides examples of different software designs that try to break filter bubbles. Secondly, we show how norms required by two democracy models dominate the tools that are developed to fight the filter bubbles, while norms of other models are completely missing in the tools. The paper in conclusion argues that democracy itself is a contested concept and points to a variety of norms. Designers of diversity enhancing tools must thus be exposed to diverse conceptions of democracy. 相似文献
18.
Social networking sites (SNSs) enable user to personalize their contents and functions. This feature has been assumed as causing positive effects on the use of online information services through enhancing user satisfaction. However, unlike other online information services (non-participatory information services), due to the results of personalization in a certain situation, SNS users cannot help using the SNS even though they feel dissatisfaction on using it. SNSs are different from other information services in the sense that they create and sustain their own value based on the number of participating members. In SNSs, personalization, reflected by updates and maintenance of profile pages, results in such participation. This study hypothesizes that personalization influences on the continued use of SNSs through two factors: switching cost (extrinsic factor) and satisfaction (intrinsic factor). Web-based survey was conducted with the samples of 677 SNS users from six universities in the US. In-person interviews were conducted with 25 university students to elicit their thoughts on the SNSs. Quantitative analysis employed by testing the proposed model with five hypotheses through a structural equation modeling (SEM) technique. The transcribed interview data was analyzed following the constant comparative technique. The main findings indicate that, as expected, the personalization increases its switching cost as well as satisfaction, which results in further use of SNSs. These findings suggest that it is necessary to consider both extrinsic and intrinsic factors of user perceptions when adding personalization features on SNSs. 相似文献
19.
20.
技术把门人在研发团队信息传递中的作用机制在互联网技术的影响下需要重新审视。对某制药企业北京研发团队进行案例研究,使用社会网络分析(SNA)并做人员访谈。分析结果表明,在互联网技术的影响下,技术把门人在研发团队中所占比例为6%,远低于以往研究公认的20%。技术把门人角色可分离为外部交流明星和内部交流明星。建立了研发团队信息传递机制新的概念框架。 相似文献