SLATE'14

Thursday, 19th June | Friday, 20th June | Keynotes

Note: Full papers have 20 minutes for presentation, plus 5 minutes for Q&A. Short papers have 10 minutes for presentation, plus 5 minutes for Q&A.

9h00	Registration
9h45	Opening Session
10h00	Keynote I: Language-driven software development José-Luis Sierra (chair: Pedro Rangel Henriques)
11h00	Coffee Break
Session I: Domain Specific Languages (chair: José-Luis Sierra)
11h30	ReCooPLa: a DSL for coordination-based reconfiguration of software architectures Flávio Rodrigues, Nuno Oliveira and Luís S. Barbosa.
11h55	A Workflow Description Language to Orchestrate Multi-Lingual Resources* Rui Brito and José João Almeida.
12h10	Converting Ontologies into DSLs* João Fonseca, Pedro Rangel Henriques and Maria João Varanda Pereira.
12h25	JSON on Mobile: is there an efficient parser?* Ricardo Queirós.
12h40	Unfuzzying Fuzzy Parsing* Pedro Carvalho, Nuno Oliveira and Pedro Henriques.
13h00	Lunch
Session II: NLP Tools and Resources (chair: Nuno Ramos Carvalho)
14h30	Automatic Identification of Proverbs Amanda Rassi, Jorge Baptista and Oto Vale.
14h55	Language Identification: a Neuronal Network approach Alberto Simões, José João Almeida and Simon Byers.
15h20	LemPORT: a High-Accuracy Cross-Platform Lemmatizer for Portuguese* Ricardo Rodrigues, Hugo Gonçalo Oliveira and Paulo Gomes.
15h35	Expanding a Database of Portuguese Tweets* Gaspar Brogueira, Fernando Batista, João Paulo Carvalho and Helena Moniz.
15h50	MLT-prealigner: a tool for multilingual text alignment* Pedro Carvalho and José João Almeida.
16h05	Coffee Break
Session III: Program Comprehension (chair: Jan Janousek)
16h30	Conclave: Writing Programs to Understand Programs Nuno Ramos Carvalho, Joé João Almeida, Maria João Varanda Pereira and Pedro Rangel Henriques.
16h55	Leveraging Program Comprehension with Concern-oriented Source Code Projections Jaroslav Porubän and Milan Nosáľ.
17h20	Comment-based concept location over system dependency graphs* Nuno Pereira, Pedro Henriques and Maria João Varanda Pereira.

Note: Full papers have 20 minutes for presentation, plus 5 minutes for Q&A. Short papers have 10 minutes for presentation, plus 5 minutes for Q&A.

9h00	Registration
9h30	Keynote II: Open Information Extraction Pablo Gamallo (chair: José João Almeida)
10h30	Coffee Break
Session IV: Semantics in NLP (chair: Pablo Gamallo)
11h00	Assigning Polarity Automatically to the Synsets of a Wordnet-like Resource Hugo Gonçalo Oliveira, António Paulo Santos and Paulo Gomes.
11h25	Detecting a tweet’s topic within a large number of Portuguese Twitter trends Hugo Rosa, Joao Paulo Carvalho and Fernando Batista.
11h50	Multiscale parameter tuning of a semantic relatedness algorithm José Paulo Leal and Teresa Costa.
12h15	Rocchio’s Model Based on Vector Space Basis Change for Pseudo Relevance Feedback Rabeb Mbarek, Mohamed Tmar and Hawete Hattab.
12h40	Automatic Identification of Whole-Part Relations in Portuguese* Ilia Markov, Nuno Mamede and Jorge Baptista.
13h00	Lunch
Session V: Programming Languages and Compilers (chair: Paulo Matos)
14h30	Contract-Java: Design by Contract in Java with Safe Error Handling Miguel Oliveira e Silva and Pedro G. Francisco.
14h55	Implementing Python for DrRacket Pedro Ramos and António Leitão.
15h20	Plagiarism detection: A tool survey and comparison Vítor T. Martins, Daniela Fonte, Pedro Rangel Henriques and Daniela Da Cruz.
15h45	Target Code Selection by Tilling AST with the Use of Tree Pattern Pushdown Automaton* Jan Janousek.
16h00	Coffee Break
Lightning Talks and Demos Session (chair: Alberto Simões)
16h30	DEMO: The LARA Language and its Compiler Suite Tiago Carvalho, Ricardo Nobre, João Bispo, Pedro Pinto, João M.P. Cardoso
16h40	DEMO: Behind PoeTryMe — a poetry generation platform Hugo Gonçalo Oliveira
16h50	DEMO: Generating LaTeX syntactic diagrams Miguel Oliveira e Silva
17h00	DEMO: A MATLAB to C/OpenCL Compiler João Bispo, Luís Reis, Ricardo Nobre, Pedro Pinto, Tiago Carvalho, João M.P. Cardoso
17h10	DEMO: EvDebugger: An Attribute Grammar Visual Debugger Daniel Rodríguez Cerezo, Pedro Rangel Henriques, José-Luis Sierra
17h30	Awards and Closing Session

Keynote I - Language-driven software development
(José-Luis Sierra from Complutense University of Madrid, Spain)

“Language-driven software development” consists in applying computer language design‑and‑implementation techniques to build conventional software. In this talk we will review two different language-driven development approaches: domain-specific languages (DLSs), and language-oriented architectures (LOAs). The DSL approach focuses on the provision of languages specialized in different application aspects, which are used by developers, and even by domain experts, during application construction and maintenance. The LOA strategy, in its turn, conceives applications themselves as coordinated collections of language processors, which can be developed using language implementation tools (parser generators, attribute grammar-based systems, etc.). In addition to presenting these approaches, we will use case studies from the fields of knowledge-based systems, e-Learning, semi-structured data processing, and Digital Humanities as illustrative examples.

About José-Luis Sierra: José-Luis Sierra got a Bs. degree on Computer Science (1991, EUI - UPM), a Ms. degree on Computer Science applied to Manufacturing (1992, UPM), and a Ms. degree and a Ph.D degree on Computer Science (1995, FI - UPM, and 2004, FI - UCM). Currently he teaches Computer Science at Complutense University of Madrid (UCM). Formerly he was with the Professor Cuena's Intelligent Systems Research Group ( ISYS), and also with the Spanish Research Council (IAI - CSIC). He leads ILSA, the research group on Language-driven Software Development at Complutense University and he is interested in the following topics: Language-driven Software Development, Attribute Grammars, Digital humanities, E-learning technologies, Markup technologies, Domain-specific languages.

Keynote II - Open Information Extraction
(Pablo Gamallo from the Universidade de Santiago de Compostela, Spain)

Building shallow semantic representations from text corpora is the first step to perform more complex semantic tasks such as text entailment, enrichment of knowledge bases, or question answering. Open Information Extraction (OIE) is a recent unsupervised strategy to extract great amounts of basic propositions, assertions, and facts (as verb-based triples) from massive corpora, which scales to Web-size document collections. The basic propositions extracted using OIE can be considered as being shallow semantic representations of those corpora.

Different strategies on OIE will be described and they will be compared with traditional Relation Extraction systems, based on a closed set of predefined relations. Finally, advantages and applications of the OIE paradigm, as well as its drawbacks and limitations will be addressed.

About Pablo Gamallo:

* short papers

Keynote I - Language-driven software development (José-Luis Sierra from Complutense University of Madrid, Spain)

Keynote II - Open Information Extraction (Pablo Gamallo from the Universidade de Santiago de Compostela, Spain)

Keynote I - Language-driven software development
(José-Luis Sierra from Complutense University of Madrid, Spain)

Keynote II - Open Information Extraction
(Pablo Gamallo from the Universidade de Santiago de Compostela, Spain)