Florin-Claudiu Pop

Service Composition Based on Natural Language Requests

The easiest way for a user to express his needs regarding a desired service is to use natural language. The main issues come from the fact that the natural language is incomplete and ambiguous, while the service composition process should lead to valid services. In this paper we propose a natural language service assemblage method based on composition templates (patterns). The use of templates assures that the composition result is always valid. The proposed system, called NLSC (Natural Language Service Composer), was implemented on the top of a service-oriented middleware called WComp and tested in an intelligent home environment.

On-demand service composition based on natural language requests

The widespread and diversity of Web services in the ubiquitous computing era and the impossibility to predict a priori all possible user demands, generates the need of a system capable of dynamically composing new services, based on unrestricted natural language requests.

Equitable solutions in QoS-aware service optimization

Web services QoS optimization problem using the concept of Lorenz dominance is addressed. Lorenz solutions are equitable and well balanced. Such an approach could simplify the Decision Makers choice. Evolutionary detection of Lorenz solution seem to be an appealing one. Some state-of-the art MOEAs are slightly modified for addressing the QoS optimization problem. Lorenz solutions drastically reduce the number of solutions in the Pareto set, and thus the decision costs.

Medical Services Optimization Using Differential Evolution

This paper proposes a method to compose and optimize medical services as business workflows. Such a workflow consists in a set of abstract services, and for each abstract service there are several concrete services. Since each medical service has different QoS (Quality of Service) parameters such as response time, rating, distance and cost, determining the optimal combination of concrete services that realize the abstract services of the business workflow is an NP hard problem. Recent proposals for solving NP optimization problems indicate the Genetic Algorithms (GA) as the best approach for complex workflows. But this problem usually needs to be solved at runtime, a task for which GA may be too slow. We propose a new approach, based on Differential Evolution (DE), that converges faster and it is more scalable and robust than the existing solutions based on Genetic Algorithms.

Natural Language Service Composition with Request Disambiguation

The aim of our research is to create a service composition system that is able to detect and deal with the imperfections of a natural language user request while keeping it as unrestricted as possible. Our solution consists in three steps: first, service prototypes are generated based on grammatical relations; second, semantic matching is used to discover actual services; third, the composed service is generated in the form of an executable plan. Experimental results have shown that inaccurately requested services can be matched in more than 95% of user queries. When missing service inputs are detected, the user is asked to provide more details.

Interactive Intonation Optimisation Using CMA-ES and DCT Parameterisation of the F0 Contour for Speech Synthesis

Expressive speech is one of the latest concerns of text-to-speech systems. Due to the subjectivity of expression and emotion realisation in speech, humans cannot objectively determine if one system is more expressive than the other. Most of the text-to-speech systems have a rather flat intonation and do not provide the option of changing the output speech. We therefore present an interactive intonation optimisation method based on the pitch contour parameterisation and evolution strategies. The Discrete Cosine Transform (DCT) is applied to the phrase level pitch contour. Then, the genome is encoded as a vector that contains 7 most significant DCT coefficients. Based on this initial individual, new speech samples are obtained using an interactive Covariance Matrix Adaptation Evolution Strategy (CMA-ES) algorithm. We evaluate a series of parameters involved in the process, such as the initial standard deviation, population size, the dynamic expansion of the pitch over the generations and the naturalness and expressivity of the resulted individuals. The results have been evaluated on a Romanian parametric-based speech synthesiser and provide the guidelines for the setup of an interactive optimisation system, in which the users can subjectively select the individual which best suits their expectations with minimum amount of fatigue.

Dynamic Service Composition for Plug-and-Play Medical Devices

Object: Clinical environments are crowded with technology where devices from multiple vendors run into interoperability barriers. Systems of integrated medical devices support improvements in workflow and reductions in medical errors and healthcare costs to the benefit of patients throughout the continuum of care.