The R+D Group MiST (MultI-paradigm Software Technology)

Project MERINET (Rigorous Methods for the Future Internet)

Summary

The MERINET project has the goal of producing new (more competitive) techniques, methods, and tools for the software analisys and development, with special emphasis in web systems and programming languages. MERINET will contribute to the Spanish industry development with (1) new research advances on software validation based on sound and rigorous techniques, and (2) new systems and tools that implement the proposed techniques to solve the current challenges of the software industry.

MERINET will focus on many different languages and formalisms, but it will focus specially on:

• HTML
• Java
• Erlang
• Prolog
• XQuery
• ATL
• SPARQL
• Bousi~Prolog
• Floper

The main areas of research in this R+D project are:

• Software testing. This is the technology most commonly used by the software industry to prevent errors during the software development. The controlled execution of a number of tests allows one to validate, to some extent, the correctness of the software developed. In this project, we will focus on automated techniques for test-case generation and for regression testing.
• Formal verification. The main drawback of the previous technique is that, despite the fact that they can be very helpful to locate the bugs of a program, they cannot be used to ensure that a program is error-free. Therefore, we will also consider formal verification techniques that finitely approximate (by means of static analysis, abstract interpretation, constraints, etc.) the possible executions of a program. Thus, we may obtain a total assurance of the correctness of a program, which makes these techniques specially adequate for validating critical software.
• Program debugging. Complementary to the previous methods, we can also find a number of debugging techniques that help the programmer to detect and locate the most common program errors. In this project, we will focus on some of the most advanced techniques, namely algorithmic debugging and program slicing.
• Information retrieval. In the area of analysis of the web, we will develope techniques for the automatic analysis, filtering, and extraction of information. Among other techniques, we will produce new algorithms for main content extraction, block detection, menu detection, and template extraction.
• Fuzzy techniques. The methods proposed for information retrieval will be complemented with fuzzy techniques that will be based on our previous experience with Bousi~Prolog and Floper.

Besides a number of scientific publications that will help us to disseminate our main achievements, we also plan to develop methods and software tools useful to improve the quality and productivity of the software development process. Therefore, the deliverables of this project will allow us to increase the international visibility of the Spanish research and will also allow the Spanish software development industry to improve its competitiveness by using the most advanced techniques for software validation.

KEY WORDS

Coordinated project

• Universidad Politécnica de Valencia (MiST group)
• Universidad de Castilla la Mancha (SMILe group)

Members of the research group (UPV subproposal)

• Carlos Herrero
• David Insa
• Marisa Llorens
• Naoki Nishida
• Javier Oliver
• Adrián Palacios
• Josep Silva
• Salvador Tamarit
• Germán Vidal (principal investigator)