Pierluigi Roberti

Process-Level composition of executable web services: ”on-the-fly” versus ”once-for-all” composition

Most of the work on automated composition of web services has focused so far on the problem of composition at the functional level, i.e., composition of atomic services that can be executed in a single request-response step. In this paper, we address the problem of automated composition at the process level, i.e., a composition that takes into account that executing a web service requires interactions that may involve different sequential, conditional, and iterative steps.We define two kinds of process-level composition problems: on-the-fly compositions that satisfy one-shot user requests specified as composition goals, and a more general form, called once-for-all compositions, whose goal is to build a general composed web service that is able to interact directly with the users, receive requests from them, and propose suitable answers. We propose a solution to these two kinds of process-level compositions, and apply the solution to the case of web services described in OWL-S. As a result, we automatically generate process-level compositions as executable OWL-S process models. We show that, while executable on-the-fly compositions can be described as standard OWL-S process models, once-for-all compositions need OWL-S process models to be extended with receive and reply constructs.

STS-tool: Socio-technical Security Requirements through social commitments

Security Requirements Engineering (SRE) deals with the elicitation and analysis of security needs to specify security requirements for the system-to-be. In previous work, we have presented STS-ml, a security requirements modelling language for Socio-Technical Systems (STSs) that elicits security needs, using a goal-oriented approach, and derives the security requirements specification based on these needs. Particularly, STS-ml relates security to the interaction among actors in the STS. In this paper, we present STS-Tool, the modelling and analysis support tool for STS-ml. STS-Tool allows designers to model a STS at a high-level of abstraction, while expressing security needs over the interactions between the actors in the STS, and derive security requirements in terms of social commitments - promises with contractual validity - once the modelling is done.

Leveraging POMDPs Trained with User Simulations and Rule-based Dialogue Management in a Spoken Dialogue System

We have developed a complete spoken dialogue framework that includes rule-based and trainable dialogue managers, speech recognition, spoken language understanding and generation modules, and a comprehensive web visualization interface. We present a spoken dialogue system based on Reinforcement Learning that goes beyond standard rule based models and computes on-line decisions of the best dialogue moves. Bridging the gap between handcrafted (e.g. rule-based) and adaptive (e.g. based on Partially Observable Markov Decision Processes - POMDP) dialogue models, this prototype is able to learn high rewarding policies in a number of dialogue situations.

STS-tool: using commitments to specify socio-technical security requirements

In this paper, we present STS-Tool, the modelling and analysis support tool for STS-ml, an actor- and goal-oriented security requirements modelling language for Socio-Technical Systems (STSs). STS-Tool allows designers to model a socio-technical system at a high-level of abstraction, while expressing constraints (security needs) over the interactions between the actors in the STS, and derive security requirements in terms of social commitments (promises with contractual validity) once the modelling is done.

Specifying and Reasoning over Socio-Technical Security Requirements with STS-Tool

We present the latest version of STS-Tool, the modelling and analysis support tool for STS-ml, an actor- and goal-oriented security requirements modelling language for socio-technical systems. STS-Tool allows designers to model a socio-technical system in terms of high-level primitives such as actor, goal, and delegation; to express security constraints over the interactions between the actors; and to derive security requirements once the modelling is done. The tool features a set of automated reasoning techniques for i checking if a given STS-ml model is well-formed, and ii determining if the specification of security requirements is consistent, that is, there are no conflicts among security requirements. These techniques have been implemented using disjuntive datalog programs. We have evaluated our tool through various industrial case studies.

Combining POMDPs trained with User Simulations and Rule-based Dialogue Management in a Spoken Dialogue System

Over several years, we have developed an approach to spoken dialogue systems that includes rule-based and trainable dialogue managers, spoken language understanding and generation modules, and a comprehensive dialogue system architecture. We present a Reinforcement Learning-based dialogue system that goes beyond standard rule-based models and computes on-line decisions of the best dialogue moves. The key concept of this work is that we bridge the gap between manually written dialog models (e.g. rule-based) and adaptive computational models such as Partially Observable Markov Decision Processes (POMDP) based dialogue managers.

Security Requirements Engineering with STS-Tool

In this chapter, we present STS-Tool, the modelling and analysis support tool for STS-ml, an actor- and goal-oriented security requirements modelling language for socio-technical systems. STS-Tool is a standalone application written in Java and based on the Eclipse RCP Framework. It supports modelling a socio-technical system in terms of high-level primitives such as actor, goal delegation, and document exchange; to express security constraints over the interactions between the actors; and to derive security requirements once the modelling is done. It also supports analysing the created STS-ml models in terms of (i) well-formedness, (ii) violation of security requirements, and (iii) threats impact over actors’ assets. We also present the architecture of STS-Tool together with its main features and provide technical details of the modelling and analysis capabilities.