Dan C. Stefanescu

Context adaptation of fuzzy systems through a multi-objective evolutionary approach based on a novel interpretability index

Context adaptation (CA) based on evolutionary algorithms is certainly a promising approach to the development of fuzzy rule-based systems (FRBSs). In CA, a context-free model is instantiated to a context-adapted FRBS so as to increase accuracy. A typical requirement in CA is that the context-adapted system maintains the same interpretability as the context-free model, a challenging constraint given that accuracy and interpretability are often conflicting objectives. Furthermore, interpretability is difficult to quantify because of its very nature of being a qualitative concept. In this paper, we first introduce a novel index based on fuzzy ordering relations in order to provide a measure of interpretability. Then, we use the proposed index and the mean square error as goals of a multi-objective evolutionary algorithm aimed at generating a set of Pareto-optimum context-adapted Mamdani-type FRBSs with different trade-offs between accuracy and interpretability. CA is obtained through the use of specifically designed operators that adjust the universe of the input and output variables, and modify the core, the support and the shape of fuzzy sets characterizing the partitions of these universes. Finally, we show results obtained by using our approach on synthetic and real data sets.

Distributed evaluation of generalized path queries

Nowadays, we are required to deal with more complex data, prime examples of which are data on the Web, XML data, biological data, etc. There are already proposed abstractions to handle these kinds of data, in particular in terms of semistructured data models. A semistructured model conceives a database essentially as a finite directed labeled graph whose nodes represent objects, and whose edges represent relationships between objects. In this paper, we focus on path queries, which are considered the basic querying mechanism for semistructured data. In essence, such queries are used to navigate, or discover paths that conform to specifications captured by regular expressions. In order to make the navigation more useful, we consider generalized path queries, in which the symbols could optionally be weighted by numbers. Such numbers can express a variety of information about the data that the query could possibly match or navigate.Motivated by the plethora of todays applications utilizing Web services and peer-to-peer architectures, we present a distributed algorithm for evaluating generalized path queries. We follow a realistic model with distributed (non-shared) memory and message-passing between processors. An optimal solution to the problem lies in the intersection of ideas related to distributed query evaluation, distributed shortest path computation, and queueing systems.

Enhanced regular path queries on semistructured databases

Regular path queries are the basic navigational component of virtually all the mechanisms for querying semistructured data commonly found in information integration applications, Web and communication networks, biological data management etc. We start by proposing weight-enhanced regular path queries with semantics that allow user-assigned preference (query) weights to be naturally combined with quantitative database link-information for driving the navigation. Motivated by the fact that the main applications of semistructured data involve distributed data sources, we focus next on the distributed evaluation of the weight-enhanced path queries. We present a distributed algorithm for evaluating our proposed queries in a multi-source setting. Our algorithm is general in the sense that it does not assume a known topology of the network and it can work using asynchronous communication. This algorithm can also be used to solve multi-source shortest path problems for which the full graph is not known in advance. To the best of our knowledge our algorithm is the first to address this problem in such a setting.

Grid-Aware Evaluation of Regular Path Queries on Spatial Networks

Regular path queries (RPQs), expressed as regular expressions over the alphabet of database edge-labels, are commonly used for guided navigation of graph databases. While convenient to use, RPQs are notorious for their high computational demand. In this paper, we present a grid- aware, fault tolerant distributed algorithm for answering RPQs on spatial networks. We engineer each part of the algorithm to account for the assumed computational-grid setting. We experimentally evaluate our algorithm, and show that for typical user queries, our algorithm satisfies the desiderata for distributed computing in general, and computational-grids in particular.

Exploiting Fuzzy Ordering Relations to Preserve Interpretability in Context Adaptation of Fuzzy Systems

In the framework of context adaptation of fuzzy systems, a typical requirement of a contextualized system is to maintain the same interpretability as the original one. Here, we propose a novel index based on a fuzzy ordering relation to provide a measure of interpretability. Our index assesses ordering, distinguishability and coverage at the same time. We use the proposed index and the mean square error as goals of a multi-objective genetic algorithm aimed at generating contextualized Mamdani fuzzy systems with different trade-offs between the two goals. Results obtained on a synthetic data set are also discussed.

Rolling element bearing diagnosis using convex hull

In this paper, we compare traditional classifiers, such as Linear and Quadratic Discriminant Classifiers and neural networks, with a one-class classifier, namely, convex hull. With reference to rolling element bearing diagnosis, we show that convex hull outperforms traditional classifiers in the classification of faults and different levels of fault severity not known during the training phase.

Segmentation and reconstruction of the lung and the mediastinum volumes in CT images

An automated system is developed for lung and mediastinum segmentation in lung CT (Computed Tomography) images for the purpose of using these segmentations not only in CT images but also in PET (Positron Emission Tomography) images to exploit the useful integration of the CT and PET images performed by the highly valuable oncological equipment PET/CT. Segmentation is the most crucial step in a CAD (Computer-Aided Diagnosis) system as lung borders delimit the region inside which pathological areas are searched for, while mediastinum borders identify the region containing the lymph nodes used for staging and restaging phases. Our method consists of an appropriate combination of image processing techniques. It was tested on CT images with different attributes such as resolution and slice thickness, containing 47 cancerous areas. We achieved almost 90% and 100% correct segmentation for lung and mediastinum, respectively.

PROD: Relayed file retrieving in overlay networks

To share and exchange the files among Internet users, peer-to-peer (P2P) applications build another layer of overlay networks on top of the Internet infrastructure. In P2P file sharing systems, a file request takes two steps. First, a routing message is generated by the client (request initiator) and spread to the overlay network. After the process finishes, the location information of the requested file is returned to the client. In the second step, the client establishes direct connection(s) with the peer(s) who store a copy of that file to start the retrieving process. While numerous research projects have been conducted to design efficient, high-performance routing algorithms, few work concentrated on file retrieving performance. In this paper, we propose a novel and efficient algorithm - PROD to improve the file retrieving performance in DHT based overlay networks. In PROD, when a file or a portion of a file is transferred from a source peer to the client, instead of creating just one direct link between these two peers, we build an application level connection chain. Along the chain, multiple network links are established. Each intermediate peer on this chain uses a store-and-forward mechanism for the data transfer. PROD also introduces a novel topological based strategy to choose these peers and guarantees the transmission delay of each intermediate link is much lower than the direct link. We conducted extensive simulation experiments and the results shown that PROD can greatly reduce the transfer time per file in DHT base P2P systems.

Semantic networks of interests in online non-suicidal self-injury communities

People who engage in non-suicidal self-injury (NSSI) often conceal their practices, which limits examination and understanding of their engagement. The goal of this research is to utilize data from public online social networks (namely, LiveJournal, a major blogging social networking site) to observe the NSSI population in a naturally occurring setting. Specifically, the focus of this paper is the interests publicly declared by LiveJournal users. In the course of study, we collected the self-declared interests of 25,000 users who are members of or participate in 139 NSSI-related communities. We constructed a family of semantic networks of interests based on their similarity. The semantic networks are structured and contain several dense clusters—semantic domains—that include NSSI-specific interests (such as self-injury and razor), references to music performers (such as evanescence), and general daily life and creativity related interests (such as poetry and friendship). Assuming users are genuine in their declarations, the clusters reveal distinct patterns of interest and may signal keys to NSSI.