Homogenization with uncertainty in Poisson ratio for polymers with rubber particles

The main aim of this work is dual computer analysis of probabilistic coefficients for the homogenized tensor of the polymer filled with the rubber particles having randomized Poisson ratios of both constituents. The major issue is to verify an influence of a randomness in rubber Poisson ratio close to the compressibility limit on the uncertainty of the effective tensor probabilistic characteristics. Probabilistic analysis presented here is carried out using mainly the stochastic perturbation technique provided by the common application of the traditional FEM commercial code ABAQUS and the symbolic computations package MAPLE. This FEM-based technique employs polynomial response function of the optimum order recovered from the weighted least squares method and following a set of deterministic solutions obtained for various values of the randomized input parameter. Optimization procedure is released entirely into a symbolic environment, where maximization of the correlation factor together with minimization of the fitting variance and approximation error are applied. Homogenization technique consists in equating of deformation energies for the real composite and the artificial one characterized by the effective elasticity tensor with uncertainty.     

A hybrid online-product recommendation system: Combining implicit rating-based collaborative filtering and sequential pattern analysis

Many online shopping malls in which explicit rating information is not available still have difficulty in providing recommendation services using collaborative filtering (CF) techniques for their users. Applying temporal purchase patterns derived from sequential pattern analysis (SPA) for recommendation services also often makes users unhappy with the inaccurate and biased results obtained by not considering individual preferences. The objective of this research is twofold. One is to derive implicit ratings so that CF can be applied to online transaction data even when no explicit rating information is available, and the other is to integrate CF and SPA for improving recommendation quality. Based on the results of several experiments that we conducted to compare the performance between ours and others, we contend that implicit rating can successfully replace explicit rating in CF and that the hybrid approach of CF and SPA is better than the individual ones.     

Pricing decisions with retail competition in a fuzzy closed-loop supply chain

The optimal pricing decision problem of a fuzzy closed-loop supply chain with retail competition is considered in this paper. The fuzziness is associated with the customer demands, the remanufacturing cost and the collecting cost. By using game theory and fuzzy theory, the optimal decision on wholesale price, retail prices and remanufacturing rate are explored respectively under the centralized and the decentralized decision scenarios, and the expressions for them are also established. Some insights into the economic behavior of firms are given, which can serve as the basis for further study in the future.     

Improved conjugate gradient implementation for least squares support vector machines

As a promising method for pattern recognition and function estimation, least squares support vector machines (LS-SVM) express the training in terms of solving a linear system instead of a quadratic programming problem as for conventional support vector machines (SVM). In this paper, by using the information provided by the equality constraint, we transform the minimization problem with a single equality constraint in LS-SVM into an unconstrained minimization problem, then propose reduced formulations for LS-SVM. By introducing this transformation, the times of using conjugate gradient (CG) method, which is a greatly time-consuming step in obtaining the numerical solution, are reduced to one instead of two as proposed by Suykens et al. (1999). The comparison on computational speed of our method with the CG method proposed by Suykens et al. and the first order and second order SMO methods on several benchmark data sets shows a reduction of training time by up to 44%.     

Site selection in waste management by the VIKOR method using linguistic assessment

Site selection is an important issue in municipal solid waste (MSW) management. Selection of the appropriate solid waste site is an extensive evaluation process that requires consideration of multiple alternative solutions and evaluation criteria. In reality, it is easier for decision makers to express their judgments on the alternatives by using linguistic terms, and there usually exists uncertain and incomplete assessment information. Moreover, decision makers may have different risk attitudes in the siting process because of their different backgrounds and personalities. Therefore, an attitudinal-based interval 2-tuple linguistic VIKOR (ITL-VIKOR) method is proposed in this paper to select the best disposal site for MSW. The feasibility and practicability of the proposed method are further demonstrated through an example of refuse-derived fuel (RDF) combustion plant location. Results show that the new approach is more suitable and effective to handle the MSW site selection problems by considering the decision maker's attitudinal character and incorporating the uncertain and incomplete assessment information.     

Fuzzy system based human behavior recognition by combining behavior prediction and recognition

With the development of intelligent surveillance systems, human behavior recognition has been extensively researched. Most of the previous methods recognized human behavior based on spatial and temporal features from (current) input image sequences, without the behavior prediction from previously recognized behaviors. Considering an example of behavior prediction, “punching” is more probable in the current frame when the previous behavior is “standing” as compared to the previous behavior being “lying down.” Nevertheless, there has been little study regarding the combination of currently recognized behavior information with behavior prediction. Therefore, we propose a fuzzy system based behavior recognition technique by combining both behavior prediction and recognition. To perform behavior recognition during daytime and nighttime, a dual camera system of visible light and thermal (far infrared light) cameras is used to capture 12 datasets including 11 different human behaviors in various surveillance environments. Experimental results along with the collected datasets and open database showed that the proposed method achieved higher accuracy of behavior recognition when compared to conventional methods.     

Consumer interest in receiving information through social media about the risks of pesticide residues

A consumer segmentation approach was used to determine consumer interest in using social media to obtain information concerning the risks of pesticide residues on vegetables. A total of 497 Flemish consumers participated in an online survey in March 2012 to assess interest in social media applications like Facebook, Twitter, YouTube, forums and blogs, and Wikipedia as channels for receiving information about pesticide residues from official bodies. The participants were segmented in four clusters depending on their level of information sufficiency and level of interest in social media applications: “very satisfied; interested” (24%), “not satisfied; very interested” (28%), “not satisfied; some interest” (25%) and “satisfied; little interest” (23%). The segments with higher levels of satisfaction displayed higher trust in oneself and in public bodies to deal with the risks of pesticide residues and also contained relatively more males. Differences in information sufficiency between the segments were associated with different methods of information acquisition, i.e. by purposeful information seeking or by incidental information acquisition. The segments with higher interest in social media displayed higher familiarity and higher appreciation for the opportunities of social media. These segments also contained relatively younger participants. Popular channels like Facebook and Twitter scored low in terms of perceived usefulness, even among high users. Wikipedia on the other hand was perceived by all segments as the most useful and credible online source when it comes to information about risks of pesticide residues. Speed and accessibility were the main motives for using social media applications as an information channel while lack of trustworthiness was the main barrier.     

An unsupervised multilingual approach for online social media topic identification

Social media data can be valuable in many ways. However, the vast amount of content shared and the linguistic variants of languages used on social media are making it very challenging for high-value topics to be identified. In this paper, we present an unsupervised multilingual approach for identifying highly relevant terms and topics from the mass of social media data. This approach combines term ranking, localised language analysis, unsupervised topic clustering and multilingual sentiment analysis to extract prominent topics through analysis of Twitter's tweets from a period of time. It is observed that each of the ranking methods tested has their strengths and weaknesses, and that our proposed ‘Joint’ ranking method is able to take advantage of the strengths of the ranking methods. This ‘Joint’ ranking method coupled with an unsupervised topic clustering model is shown to have the potential to discover topics of interest or concern to a local community. Practically, being able to do so may help decision makers to gauge the true opinions or concerns on the ground. Theoretically, the research is significant as it shows how an unsupervised online topic identification approach can be designed without much manual annotation effort, which may have great implications for future development of expert and intelligent systems.     

Multi-level abstraction for trace comparison and process discovery

Many information systems record executed process instances in the event log, a very rich source of information for several process management tasks, like process mining and trace comparison. In this paper, we present a framework, able to convert activities in the event log into higher level concepts, at different levels of abstraction, on the basis of domain knowledge. Our abstraction mechanism manages non trivial situations, such as interleaved activities or delays between two activities that abstract to the same concept.Abstracted traces can then be provided as an input to an intelligent system, meant to implement a variety of process management tasks, significantly enhancing the quality and the usefulness of its output.In particular, in the paper we demonstrate how trace abstraction can impact on the quality of process discovery, showing that it is possible to obtain more readable and understandable process models.We also prove, through our experimental results, the impact of our approach on the capability of trace comparison and clustering (realized by means of a metric able to take into account abstraction phase penalties) to highlight (in)correct behaviors, abstracting from details.     

Face recognition using point symmetry distance-based RBF network

In this paper, a face recognition technique using a radial basis function neural network (RBFNN) is presented. The centers of the hidden layer units of the RBFNN are selected by using a heuristic approach and point symmetry distance as similarity measure. The performance of the present method has been evaluated using the AT&T Laboratories Cambridge database (formerly called ORL face database) and compared with some other methods, which use the same database. The evaluation has been done using two methodologies; first with no rejection criteria, and then with rejection criteria. The experimental results show that the present method achieves excellent performance, both in terms of recognition rates and learning efficiency. The average recognition rates, as obtained using 10 different permutations of 1, 3 and 5 training images per subject are 76.06, 92.61 and 97.20%, respectively, when tested without any rejection criteria. On the other hand, by imposing rejection criteria, the average recognition rates of the system become 99.34, 99.80 and 99.93%, respectively, for the above permutations of the training images. The system recognizes a face within about 22 ms on a low-cost computing system with a 450 MHz P-III processor, and thereby extending its capability to identify faces in interframe periods of video and in real time.