Informatica

Fuzzy c-means (FCM) is a well-known and widely applied fuzzy clustering method. Although there have been considerable studies which focused on the selection of better fuzzifier values in FCM, there is still not one widely accepted criterion. Also, in practical applications, the distributions of many data sets are not uniform. Hence, it is necessary to understand the impact of cluster size distribution on the selection of fuzzifier value. In this paper, the coefficient of variation (CV) is used to measure the variation of cluster sizes in a data set, and the difference of coefficient of variation (DCV) is the change of variation in cluster sizes after FCM clustering. Then, considering that the fuzzifier value with which FCM clustering produces minor change in cluster variation is better, a criterion for fuzzifier selection in FCM is presented from cluster size distribution perspective, followed by a fuzzifier selection algorithm called CSD-m (cluster size distribution for fuzzifier selection) algorithm. Also, we developed an indicator called Influence Coefficient of Fuzzifier ($\mathit{ICF}$) to measure the influence of fuzzifier values on FCM clustering results. Finally, experimental results on 8 synthetic data sets and 4 real-world data sets illustrate the effectiveness of the proposed criterion and CSD-m algorithm. The results also demonstrate that the widely used fuzzifier value $m=2$ is not optimal for many data sets with large variation in cluster sizes. Based on the relationship between ${\mathit{CV}_{0}}$ and $\mathit{ICF}$, we further found that there is a linear correlation between the extent of fuzzifier value influence and the original cluster size distributions.

This paper studies a set of novel integrated scheduling problems by taking into account the combinatorial features of various groups, parallel-batching, deteriorating jobs, and time-dependent setup time simultaneously under the settings of both single-machine and parallel-machine, and the objective of the studied problems is to minimize the makespan. In order to solve the single-machine scheduling problem, we first investigate the structural properties on jobs sequencing, jobs batching, and batches sequencing for the optimal solution, and then develop a scheduling rule. Moreover, for solving the parallel-machine scheduling problem, we exploit the optimal structural properties and batching rule, and propose a novel hybrid AIS-VNS algorithm incorporating Artificial Immune System algorithm (AIS) and Variable Neighbourhood Search (VNS). Extensive computational experiments are conducted to evaluate the performance of the proposed AIS-VNS algorithm, and comparison results show that the proposed algorithm performs quite well in terms of both efficiency and solution quality.

In this paper, we propose a novel trust inference framework in the web-based scenarios which are assumed to have a Web of Trust pre-established, and take the contexts of the trust relationships into account when inferring the recommendation trust. For alleviating the problem of sparse matrix in the Web of Trust, we also incorporate the users’ profile and relationship information on the associated social networks into the framework. Based on the Web of Trust established in the discussed web-based scenario (i.e. epinions.com in this paper), and the social relationship information in the associated social networks, the users are classified into four classes. Then different information is used to infer the users’ recommendation trust value based on the classifications. The simulation experiments show that our approach has good coverage of inferred trust values, and the accurate rate of the predicted trust relationship is higher than the traditional PCC (Pearson Correlation Co-efficiency). According to the computation results of adjusted parameters, it can be concluded that the threshold which is used to filter the inferred trust values can be removed, i.e. all the inferred trust values should be kept.