Shahab Mokarizadeh

Ontology learning for cost-effective large-scale semantic annotation of web service interfaces

In this paper we introduce a novel unsupervised ontology learning approach, which can be used to automatically derive a reference ontology from a corpus of web services for annotating semantically the Web services in the absence of a core ontology. Our approach relies on shallow parsing technique from natural language processing in order to identify grammatical patterns of web service message element/part names and exploit them in construction of the ontology. The generated ontology is further enriched by introducing relationships between similar concepts. The experimental results on a set of global Web services indicate that the proposed ontology learning approach generates an ontology, which can be used to automatically annotate around 52% of element part and field names in a large corpus of heterogeneous Web services.

Applying Semantic Web Service Composition for Action Planning in Multi-robot Systems

In this paper we demonstrate how the Web services based solutions can be effectively utilized and integrated into robotic world. In particular, we consider robotic systems where overall control is not embedded into any of the robots and the local behavior of each robot is loosely dependent on behavior of other robots. We propose an architecture for swarm action planning based on Web services paradigm exploiting a problem ontology for service discovery, linear logic based service composition for action planning and a task allocation layer for finding the most suitable robot to perform an action. In our solution all entities in the system expose their functionalities as Web services and allow dynamic service discovery and selection.

Evaluation of a Semi-automated Semantic Annotation Approach for Bootstrapping the Analysis of Large-Scale Web Service Networks

In recent years many methods have been proposed, which require semantic annotations of Web services as an input. Such methods include discovery, match-making, composition and execution of Web services in dynamic settings, just to mention few. At the same time automated Web service annotation approaches have been proposed for supporting application of former methods in settings where it is not feasible to provide the annotations manually. However, lack of effective automated evaluation frameworks has seriously limited proper evaluation of the constructed annotations in practical settings where the overall annotation quality of millions of Web services needs to be evaluated. This paper describes an evaluation framework for measuring the quality of semantic annotations of large number of Web services descriptions provided in form of WSDL and XSD documents. The evaluation framework is based on analyzing network properties, namely scale-free and small-world properties, of Web service networks, which in turn have been constructed from semantic annotations of Web services. The evaluation approach is demonstrated through evaluation of a semi-automated annotation approach, which was applied to a set of publicly available WSDL documents describing altogether ca 200 000 Web service operations.

Trust and Privacy Enabled Service Composition using Social Experience

In this paper, we present a framework for automatic selection and composition of services which exploits trustworthiness of services as a metric for measuring the quality of service composition. Trustworthiness is defined in terms of service reputation extracted from user profiles. The profiles are, in particular, extracted and inferred from a social network which accumulates users past experience with corresponding services. Using our privacy inference model we, first, prune social network to hide privacy sensitive contents and, then, utilize a trust inference based algorithm to measure reputation score of each individual service, and subsequently trustworthiness of their composition.

Epidemic Trust-Based Recommender Systems

Collaborative filtering(CF) recommender systems are among the most popular approaches to solving the information overload problem in social networks by generating accurate predictions based on the ratings of similar users. Traditional CF recommenders suffer from lack of scalability while decentralized CF recommenders (DHT-based, Gossip-based etc.) have promised to alleviate this problem. Thus, in this paper we propose a decentralized approach to CF recommender systems that uses the T-Man algorithm to create and maintain an overlay network that in turn would facilitate the generation of recommendations based on local information of a node. We analyse the influence of the number of rounds and neighbors on the accuracy of prediction and item coverage and we propose a new approach to inferring trust values between a user and its neighbors. Our experiment son two datasets show an improvement of prediction accuracy relative to previous approaches while using a highly scalable, decentralized paradigm. We also analyse item coverage and show that our system is able to generate predictions for significant fraction of the users, which is comparable with the centralized approaches.

A rule-based semantic matching of base object models

Creating simulation models via composition of predefined and reusable components is an efficient way of reducing costs and time associated with the simulation model development. However, to successfully compose models one has to solve the issues of syntactic and semantic composability of components. The Base Object Model (BOM) standard is an attempt to ease reusability and composition of simulation models. However, the BOM does not contain sufficient information for defining necessary concepts and terms to avoid ambiguity, and neither does it have any method for dynamic aspects matching conceptual models (i.e., their state-machines). In this paper, we present our approach for enhancement of the semantic contents of BOMs and propose a three-layer model for syntactic and semantic matching of BOMs. The enhancement includes ontologies for entities, events and interactions in each component. We also present an OWL-S description for each component, including the state-machines. To test our approach, we specify some simulation scenarios and implement BOMs as building blocks for development of those scenarios, one of which is presented in this paper. We also define composability degree, which quantifies closeness of the composed model to a given model specification. Our results show that the three-layer model is promising and can improve and simplify the composition of BOM-based components.

Information Diffusion in Web Services Networks

Information diffusion has been studied between and within biosphere, microblogs, social networks, citation networks and other domains, where the network structure is present. These studies have turned to be useful for acquiring intrinsic knowledge for strategic decision making in related areas, for example, planning online campaigns in case of microblogs and blogosphere. In the context of data-centric Web services, information exchange patterns will reveal practical heuristics for efficient Web services selection and composition. For example, by possible knowledge that there is information flow between Web services of \textit{Financial Analysis Services} and \textit{Enterprise Resource Planning Services}, as outlined by our experimental results, the potential applications, which interact with \textit{Financial Analysis} services, can be adjusted to take advantage of \textit{Enterprise Resource Planning} services as well. In this paper we present a method for analyzing information diffusion between categories of data-centric Web services. The method operates on a Web services network constructed by linking interface descriptions of categorized Web services. The proposed method is evaluated on a case study of global Web services. The results indicate high potential of the proposed model in understanding interactions between categories of Web services.

Utilizing Web Services Networks for Web Service Innovation

The increasing presence and adoption of Web services on the Web has promoted the significance of management of new service development for service developing sectors. The major challenge is that how to find missing but potentially valuable Web services to be developed. This problem can be divided into two sub-problems: finding missing Web services and measuring the added-value of the introduced services. This paper addresses a plausible solution to the first sub problem. Given a collection of Web services, we propose a framework for suggesting a set of candidate Web services that can be introduced to the collection. These suggested services are novel and do not present in the given collection. Our solution relies on the network structure of Web services for finding and recommending new Web services and utilizes the already observed properties of Web services networks for collective evaluation of the suggested services. The proposed solution is evaluated using 753 semantically annotated Web services. The experimental results shows that the proposed framework provides web service community with new network driven methods for finding and evaluation of new Web services.

Ontology acquisition from web service descriptions

The lack of formally expressed semantics in web services complemented with the increasing number of available web services is the main obstacle in analyzing and using the existing web services exposed in the Web. In the absence of appropriate reference domain ontology, annotation of existing web services is dependent on ontology development and ontology learning techniques. In this paper we present an unsupervised ontology learning approach tailored to learning from WSDL documents. The most specific feature of the suggested approach is that it constructs (semi-) automatically ontology fragments from a collection of WSDL documents, that lack any extra textual documentation, by just exploiting element names in the WSDL document. The suggested approach combines both linguistic and statistic analysis techniques such as lexico-syntactic patterns and term co-occurrence analysis. The preliminary results show that the generated ontology captures correctly more than half of the semantic classes and instances as well as taxonomic and non-taxonomic relations, hence, providing a reasonable basis for automatic web services annotation.

Exploiting Dynamic Privacy in Socially Regularized Recommenders

In this paper we introduce a privacy-aware collaborative filtering recommender framework which aims to address the privacy concern of profile owners in the context of social trust sparsity. While sparsity in social trust is mitigated by similarity driven trust using a probabilistic matrix factorization technique, the privacy issue is addressed by employing a dynamic privacy inference model. The privacy inference model exploits the underlying inter-entity trust information to obtain a personalized privacy view for each individual in the social network. We evaluate the proposed framework by employing an off-the-shelf collaborative filtering recommender method to make predictions using this personalized view. Experimental results show that our method offers better performance than similar non-privacy aware approaches, while at the same time meeting user privacy concerns.