Search

Zero Defects is one of the ultimate targets for manufacturing quality control and assurance. Such systems are becoming common in advanced manufacturing industries but are at an initial stage in more traditional industrial sectors, such as wood panels, laminates production, pulp and paper processing and composite panels production. This paper proposes the PREFAB framework, applied to the wood-based panels industry, to minimize rejected products using AI, machine learning and IoT devices. The framework was built through action research with a Portuguese wood-based panel manufacturing. This framework delivered an innovative decision support system that provides relevant and timely recommendations for shopfloor decision making and to support process/product engineering.

Spoken European Portuguese (EP) is known to be difficult to understand for L2 learners, due to phenomena such as strong vowel reduction. In this paper, we present a method to automatically generate exercises aimed at improving listening comprehension skills in EP. Learners identify the words pronounced in real speech utterances. The exercises introduce two innovative aspects: using broadcast news videos for curriculum and automatically generating exercises with material updated on a daily basis. The videos are automatically transcribed by a speech recognition engine. A filtering chain, used to select appropriate sentences, was validated by a first survey comprised of both manually and automatically selected sentences. Both sets were assigned good to very good subjective quality scores. A second survey concerned the features of the exercise interface. Subjects with varying self-reported exposure to Portuguese as a second language tested several interfaces and functionalities and highlighted their preferred features. The results confirmed that the largest difficulty was the fast speech rate. All participants valued slowed-down audio and video documents, though this feature was more often used by the lowest proficiency subjects. The exercises were integrated into a Web platform where they are automatically updated daily. Though further evaluation is needed to find whether the platform affords skill acquisition, it is expected to be particularly valuable for distance learners who need opportunities to access authentic audio documents in EP.

This paper targets an understanding of how metadiscourse functions in spoken language. Starting from a metadiscourse taxonomy, a set of TED talks is annotated via crowdsourcing and then a lexical grade level predictor is used to map the distri- bution of the distinct discourse functions of the taxonomy across levels. The paper concludes showing how speakers use these functions in presentational settings.

UID/LIN/03213/2013 SFRH/BD/85737/2012 SFRH/BD/51157/2010 SFRH/BD/51156/2010 A detailed error analysis is a fundamental step in every natural lan- guage processing task, as to be able to diagnosis what went wrong will provide cues to decide which are the research directions to be followed. In this paper we focus on error analysis in Machine Translation. We deeply extend previous error taxonomies so that translation errors associated with Romance languages speci- ficities can be accommodated. Also, based on the proposed taxonomy, we carry out an extensive analysis of the errors generated by four di↵erent systems: two mainstream online translation systems Google Translate (Statistical) and Systran (Hybrid Machine Translation) and two in-house Machine Translation systems, in three scenarios representing di↵erent challenges in the translation from English to European Portuguese. Additionally, we comment on how distinct error types di↵erently impact translation quality. publishersversion published

This paper describes the supervised classification of four metadiscursive functions in English. Training data is collected using crowdsourcing to label a corpus of TED talks transcripts with occurrences of Introductions, Conclusions, Examples, and Emphasis. Using decision trees and lexical features, we report classification accuracy.

Fill-in-the-blank questions are one of the main assessment devices in REAP.PT tutoring system. The problem of automatically generating the stems, i.e. the sentences that serve as basis to this type of question, has been studied mostly for English, and it remains a challenge for a language as morphologically rich as European Portuguese (EP), for which additional data scarcity problems arise. To address this problem, a supervised classification technique is used to model a classifier that decides whether a given sentence is suitable to be used as a stem in a cloze question. The major focus is put in the feature engineering task, describing both the development of new criteria, and the adaptation to EP of features already explored in the literature. The resulting classifier filters out inadequate stems, allowing experts to build and personalize their instruction focusing on a set of potentially good sentences.

This paper describes the integration of multimedia documents in the Portuguese version of REAP, a tutoring system for vocabulary learning. The documents result from the pipeline processing of Broadcast News videos that automatically segments the audio files, transcribes them, adds punctuation and capitalization, and breaks them into stories classified by topics. The integration of these materials in REAP was done in a way that tries to decrease the impact of potential errors of the automatic chain in the learning process.

In this paper, we report on the development of a methodology for the evolution of software towards new architectures. In our approach, we represent source code as graphs. This enables the use of graph transformation rules, allowing the automation of the transformation process. Prior to its model representation, the source code is subject to a preparatory step of semi-automatic code annotation according to the contribution of each of its parts in the target architecture. This paper first describes the overall methodology and then focuses on the code annotation and model transformation parts. We also discuss issues of the implementation of the approach based on existing tools.