Search

Abstract Process model matching refers to the automatic identification of corresponding activities between two process models. It represents the basis for many advanced process model analysis techniques such as the identification of similar process parts or process model search. A central problem is how to evaluate the performance of process model matching techniques. Current evaluation methods require a binary gold standard that clearly defines which correspondences are correct. The problem is that often not even humans can agree on a set of correct correspondences. Hence, evaluating the performance of matching techniques based on a binary gold standard does not take the true complexity of the matching problem into account and does not fairly assess the capabilities of a matching technique. In this paper, we propose a novel evaluation procedure for process model matching techniques. In particular, we build on the assessments of multiple annotators to define the notion of a non-binary gold standard. In this way, we avoid the problem of agreeing on a single set of correct correspondences. Based on this non-binary gold standard, we introduce probabilistic versions of precision, recall, and F-measure as well as a distance-based performance measure. We use a dataset from the Process Model Matching Contest 2015 and a total of 16 matching systems to assess and compare the insights that can be obtained by using our evaluation procedure. We find that our probabilistic evaluation procedure allows us to gain more detailed insights into the performance of matching systems than a traditional evaluation based on a binary gold standard. [ABSTRACT FROM AUTHOR]

Biological pathways have played an important role in understanding cell activities and evolution. In order to find these pathways, it is necessary to orient protein–protein interactions, which are usually given in forms of undirected networks or graphs. Previous findings indicate that orienting protein interactions can improve the process of pathway discovery. However, assigning orientation for protein interactions is a combinatorial optimization problem which has been proved to be NP-hard, making it critical to develop efficient algorithms. This paper proposes a method for orienting protein–protein interaction networks (PPIs) and discovering pathways. For our proposal, the mathematical model of the problem is given and then a genetic algorithm is designed to find the solution for the problem taking into account the problem's characteristics. We conducted multiple runs on the data of yeast PPI networks to test the best option for the problem. The obtained results were compared with a well-known algorithm (ROLS), which was shown to be the best in dealing with this problem, in terms of the run time, fitness function values, and especially the ratio of matching gold standard pathways. The results show the good performance of our approach in addressing this problem. [ABSTRACT FROM AUTHOR]

Abstract: In order to query and retrieve the rich and useful information hidden in the Deep Web efficiently, extensive research on domain-specific Deep Web Data Integration Systems (DWDIS) has been carried out in recent years. In DWDIS, large-scale automatic integration of query interfaces of domain-specific Web Databases (WDBs) remains a serious challenge due to the scale of the problem and the great diversity of the WDBs' query interfaces. To address this challenge, in this paper, we first give a definition of the constraint matrix which can accurately describe three types of constraints (hierarchical constraints, group constraints and precedence constraints) and the strengths of attributes of a query interface, and then prove that the schema tree of the query interface corresponds to only one constraint matrix, and vice versa. Furthermore, we transform the problem of integrating domain-specific query interfaces into a problem of integrating the constraint matrices and set up a multi-objective optimization problem model. To effectively solve the optimization model, some strategies to extend and merge the constraint matrices are designed. A method for automatically detecting and filtering abnormal data (noises) in the query interfaces is also proposed. More importantly, a novel and efficient algorithm applicable to large-scale automatic integration of domain-specific query interfaces is developed. Finally, the proposed algorithm is evaluated by experiments on the real query interface data set. Our theoretical analysis and experimental results show that the proposed algorithm outperforms existing state-of-the-art integration algorithms of domain-specific query interfaces. [Copyright &y& Elsevier]

Personalisation of web information systems (WISs) means customisation of the presented data content to the needs of users, restricting the available functionality to the goals and preferences of users, and tailoring the web presentation according to used devices and style options. This paper primarily concentrates on the customisation of functionality by making all those operations available to a user that are needed to achieve a specified goal, and by organising them in an action scheme called plot that is in accordance with the behavioural preferences of the user. Plots are formalised by algebraic expressions in Kleene algebras with tests (KATs). Then personalisation can be formalised as an optimisation problem with equational preference rules, for which a term rewriting approach is proposed. In a second step the approach is extended to conditional term rewriting thereby dispensing with the particular need to associate preference rules with user profiles. Finally, the approach is refined by taking content specifications via extended views and abstract programs on these views into account. This leads us to reformulating the personalisation problem in higher-order dynamic logic. [Copyright &y& Elsevier]

In this paper, a novel robust deterministic annealing (RDA) algorithm is developed for data clustering. This method takes advantage of conventional noise clustering (NC) and deterministic annealing (DA) algorithms in terms of the independence of data initialization, the ability to avoid poor local optima, the better performance for unbalanced data, and the robustness against noise and outliers. In addition, a cluster validity criterion, i.e., Vapnik–Chervonenkis (VC)-bound induced index, which is estimated based on the structural risk minimization (SRM) principle, is specifically extended for RDA to determine the optimal cluster number for a given data set. The superiority of the proposed RDA clustering algorithm is supported by experimental results. [Copyright &y& Elsevier]