Abstract:Similarity metric of process model is an indispensable task in business process management, which is widely used in many scenarios such as organizations merging, user requirements change, and model repository management. This paper focuses on the similarity metric algorithm based on the principal transition sequences (PTS) and puts forward improvement scheme by defining the complete firing sequences to express model behavior, matching fining sequences by A^* algorithm with pruning strategy and proposing new similarity metric method. Experiments show that this method improves rationality than existing behavior-based similarity algorithms.

Abstract:It is common for large enterprises or organizations to maintain repositories of process models. A large number of process model variants may exist in these repositories due to the differences of modeling objective or scenario, customization or tailoring reference models, and model updating modification. This paper focuses on the study of identifying commonalities and differences between process variants, i.e. construction of matching relations between process variants automatically. To support the discovery of complex correspondences and ensure the effectiveness and efficiency of matching results, we propose a matching technique based on the traversal of process structure tree and present a process similarity measuring method based on tree-edit distance. The experimental evaluation based on real-world process model collection shows that an effective precision and recall is achieved.

Abstract:Comparing processes has important application value in business process management, and it is common in scenarios of process version control, and process similarity calculation, process merging. Mapping elements between process models is the first step for process comparison. Current comparison solutions, however, only consider the mapping between tasks but ignore the correspondence of other elements in the process. Hence, the reliability of the results from the process comparison cannot be guaranteed. To address this issue, this paper first proposes a Petri net based model for element mapping between processes, and focuses on description of the context that is required for places mapping. Then, based on the place mapping context, a mapping algorithm is also presented. For any pair of processes modeled by Petri net, the algorithm adopts bilateral and unilateral mapping strategies where the pair of places that shares the most similar contexts is given high priority to be selected as the mapping result. The extensive experiments based on the real world data sets show the effectiveness of the proposed method as well as its efficiency in the scenario of process similarity search.

Abstract:Behavior conformance checking is a critical issue after process modeling and process running to ensure the correctness and stability of artifact-centric business process model. With the popularization of data-centric design, researches on checking data manipulation of business process becomes more important. This paper proposes a method for checking behavior conformance according to the lifecycle of artifact. First, the artifact behavior can be defined as the total order artifact snapshot sequences. The total order artifact snapshot sequences not only reflect the path of services, but also describe the change of attribute assignment states of the artifact. Next, the problem for behavior conformance checking is proved to be decidable by transforming the problem into the decidability problem of language. A Turing machine is designed as verification model. The model not only measures conformance of service path in the lifecycle, but also evaluates the correctness of attribute assignment of artifact. Further, the calculation method of fitness metric utilizing equivalent conversion of service-snapshot correlation matrix is presented. Lastly, the instantiation analysis and experiments demonstrate the correctness and validity of this method.

Abstract:Logs used in conformance checking with process models are often the event logs. Conformity between the model and the log is often measured by counting the traces which could be reconstructed and the tasks which would be evoked but were not in the running trace through rerunning the model according to the task traces in the log. However the method is not sufficiently comprehensive. While checking the model consisting of many selections with its Event Log, the conformity will be very low due to the large number of evoked tasks that are not in the running task trace. Moreover, while checking the model mainly composed by parallel branches with the log only containing sequential task traces and sharing the same task set with the model, the conformity will be very high due to the fact that only a few tasks can't be executed normally while monitoring the real behavior. To overcome the weakness of the original method, a bidirectional checking method made up of checking the accuracy of the model and checking the completeness of the log, and a new kind of log named Token Log which can describe the property of its corresponding model, are proposed in this paper. With the Token Log, the new method for conformance checking is clearer, more concise and more accurate.

Abstract:Heuristic process mining algorithm has a significant advantage in dealing with noise and incomplete logs. However, existing heuristic process mining algorithms cannot handle long-distance dependencies and lenth-2-loop structures correctly in some special situations. Besides, none of them are parallelized. To address the problems, process models are divided into multiple case models according to executed activity set at first. Then the C-nets corresponding to case models are discovered with an improved heuristic process mining algorithm in parallel. After that, these C-nets are integrated to derive the complete process model. Meanwhile, the definition of long- distance dependencies is extended to non-local dependencies between two activity sets in decision points. In addition, a more accurate long- distance dependency metrics and its corresponding mining algorithm are presented. These improvements make the proposed algorithm more accurate and efficient.

Abstract:Mining business process models from running logs is in its ascendant. Inevitably, the ever changing operational environment makes these log records diverse. Considering every mining algorithm has its pros and cons, this paper focuses on the challenge to apply a best mining algorithm against diverse logs. A novel approach, SoFi (survival of fittest integrator), is proposed to mine business process models effectively in such a diverse environment. SoFi tackles the diversity issue by utilizing domain knowledge to classify the cases in a log and applying various mining algorithms on these categories to obtain comprehensive process models as candidates for optimization. A genetic algorithm (GA) based optimizer takes these candidates as initial population for purpose of both genetic quality as well as genetic diversity. Under the principle of survival of fittest, the GA optimizer can aggregate best process fragments with context into the final process model for the entire log. Experiments on synthetic data and real cases from a telecommunication firm demonstrate the effectiveness of SoFi and comprehensive quality of mined process models in terms of replay fitness, accuracy, generalization, and simplicity.

Abstract:Task assignment strategy has a great impact on the performance of the workflow management system. The instability of human resource brings challenges to task assignment. General task assignment strategies have some deficiencies. First, they only consider the individual attributes of candidate resources, ignoring the influences to the candidate resources from other resources in process. In addition, they need to setup a capability index of each resource in advance. However, it is hard to make the capability index fit the actual situation, and a wrong capability index will make the workflow engine assign the task to the unsuitable resource, degrading the performance of workflow management system. To overcome the above deficiencies, four Q-learning-based task assignment algorithms are proposed according to different state transition views and different reward functions. Simulation experiments show that Q-learning-based task assignment algorithms can work well even without setting up a capability index in advance. Also due to their support to consider the social relationship, the average time of case completion decreases.

Abstract:To meet the needs of instance aspect handling in practical workflow applications, a model for instance aspect handling- oriented optimal scheduling of multiple activity instances is constructed. An algorithm for such scheduling optimization is presented correspondingly. It utlizes the theory of ant colony optimization to achieving the objectives of minimum acitity instances' total dwelling time and minimum acitity instances' total cost with constraints. The conception of wasted grouping time and wasted grouping cost are introduced according to the two optimization objectives, based on which the heuristic information for the ants are designed. The result of simulation experiment shows its effectiveness.

Abstract:This paper introduces a methodology towards enabling business process modeling with geographic and geospatial information. First, a comprehensive framework of geospatial constraints, formulated as a UML-based semantic model, is proposed. Next, a business process modeling language (LAWF-net) is designed to combine traditional control-flow constructs with the aforementioned geospatial constraints. To enable the execution of such models, a mapping to Coloured Petri Nets (CPN) is formulated. The proposed approach is implemented in the form of a CPN tool extension, and a case study is presented to show that the new approach is feasible in practice and can be combined and integrated with existing GIS systems.

Abstract:Mobile opportunistic networks (MONs) utilize the communication opportunities arising from node contacts to forward data in a hop-by-hop manner, and they play an important role in implementing the intensive perception and ubiquitous interconnection in Internet of Things. Opportunistic routing is the basic method for data communication in intermittently connected scenarios, so it is worth studying and has captured great interests from researchers. First, this paper introduces the concept, system architecture, typical applications of MONs, and some technical challenges in this field. Second, it reviews the opportunistic routing problem from three aspects: 1) evaluation metrics, 2) design requirements, 3) forwarding schemes. Some of the latest progresses of this research are also presented. Finally, the future research trends of opportunistic routing are analyzed and prospected.

Abstract:This paper provides a state-of-the-art survey of location privacy-preserving techniques in WSNs. First, the network model, the attack model and the performance evaluation model are reviewed. Then, existing work is classified into four types, including path camouflage, entrapment attracting, network anonymity and communication control. Further, the key mechanisms of typical location privacy-preserving protocols are elaborated. Performance analysis and comparison show that all these four strategies affect communication and energy efficiency in some degree. In addition, the path camouflage strategy mainly aims at hop-by-hop trace attack, the network anonymity strategy aims at ID analysis attack, while the entrapment attraction and communication control strategies are capable of resisting multiple types of attacks. Finally, suggestions for future research are provided.

Abstract:In unstructured P2P networks, how to rapidly and precisely locate user required resources is currently a hot issue, and is also one of core problems faced by P2P application fields. Related unstructured P2P resources location algorithms can not be optimized at the same time in respect to precision ratio, recall ration and query costs, which can cause serious network bandwidth burden and huge index maintenance costs. To address the problem, this paper proposes a query strategy called user requirement resource location strategy (U2RLS). The innovation of this strategy is to integrate user requirements, user preferences and user interest based on the original unstructured P2P network resources location flooding algorithm. The strategy subnets the user resources, and adopts the flooding mechanism and query index mechanism with user required information to locate the resources accurately. This strategy effectively avoids the netstorm caused by mass information, data overlapping, and resource search partial coverage phenomenon, so as to solve the problem that query nodes use relay nodes blindly. The experimental results show that U2RLS has high search success rate, limited network resources consumption and short response time in query process, and therefore can significantly improve the efficiency for user resource location.

Abstract:Internet background radiation (IBR) is a type of unproductive traffic which has been used for years in the network security and management fields. Traditionally, IBR can be obtained by darknets. Nevertheless, the deployment of darknets typically requires large dark address blocks which are hard to acquire and also potentially detectable and avoidable. To address the issue, this article proposes an algorithm to extract IBR from raw traffic in live networks. The algorithm is based on the notions of grey spaces, one-way flows and behavior learning and has a better performance than previous work. On one hand, the algorithm obtains IBR destined to both inactive addresses and active addresses, resulting a lower missing rate compared with algorithms based on inactive addresses. On the other hand, the algorithm employs a behavior learning mechanism. Although the metric "recall" decreases slightly, "precision" increases from about 93% to above 99% in contrast to algorithms based on one-way flows. After applying the algorithm to a live network consisting of about 1.28 million IP addresses, the study analyzes the extracted IBR from several aspects. Results show that more than 70% of the inbound flows are IBR flows in the past two years' data samples and this should draw enough attention from related research. Finally, several cases suggest the important role the live networks' IBR traffic plays in the network security and management fields.

Abstract:With the rise of the video surveillance system based on cloud computing (hereinafter referred to as the cloud video surveillance system), its complex energy consumption problems, brought by the terminal facilities, physical servers and frequent network- transmissions, can't be ignored. In this paper, the architecture and mechanism of energy consumption optimization of the system are introduced. Then, the energy consumption optimization researches are categorized into three levels: monitoring node, computing node and storage node. Next, considering the existing energy optimization theories and methods applied to the sensor networks and the generalized cloud computing data center, the energy consumption optimization methods for cloud video surveillance system are analyzed and compared with respect to the upper three levels. Finally, several key problems and future research directions for reducing the comprehensive energy consumption of the system are discussed.

Abstract:This paper first construct 2D principal manifold of the 3D point cloud which is typically unoriented and unevenly distributed in space, and give the quadratic optimized approximation of principal manifold in form of watertight mesh with spherical homeomorphism. By this method, the shape description of 3D point cloud is converted into the description of 2D principal manifold evenly spread in spherical parameter field. Then, it applies translation, rotation and scaling to the quadratic optimized mesh to align the mesh polar axis, denoting this process as initial rough alignment. Finally, the ICNP (iterative closest normal point) algorithm is used to iteratively refine the rigid transformation to bring the two meshes into the best alignment with respect to the least mean square error, and the alignment error is recorded as difference distance between two 3D point clouds. The experimental results show that the proposed 3D point cloud shape description based on quadratic optimized approximation of 2Dprincipal manifold is robust to noise and resolution, and can be used as the shape descriptor for 3D retrieval.