Sentence complexity assessment is a relatively new task in Natural Language Processing. One of its aims is to highlight in a text which sentences are more complex to support the simplification of contents for a target audience (e.g., children, cognitively impaired users, non-native speakers and low-literacy readers (Scarton and Specia, 2018)). This task is evaluated using datasets of pairs of aligned sentences including the complex and simple version of the same sentence. For Brazilian Portuguese, the task was addressed by (Leal et al., 2018), who set up the first dataset to evaluate the task in this language, reaching 87.8% of accuracy with linguistic features. The present work advances these results, using models inspired by (Gonzalez-Garduño and Søgaard, 2018), which hold the state-of-the-art for the English language, with multi-task learning and eye-tracking measures. First-Pass Duration, Total Regression Duration and Total Fixation Duration were used in two moments; first to select a subset of linguistic features and then as an auxiliary task in the multi-task and sequential learning models. The best model proposed here reaches the new state-of-the-art for Portuguese with 97.5% accuracy 1 , an increase of almost 10 points compared to the best previous results, in addition to proposing improvements in the public dataset after analysing the errors of our best model.

Most research on Lexical Simplification (LS) addresses non-native speakers of English, since they are numerous and easy to recruit. This makes it difficult to create LS solutions for other languages and target audiences. This paper presents SIMPLEX-PB 2.0, a dataset for LS in Brazilian Portuguese that, unlike its predecessor SIMPLEX-PB, accurately captures the needs of Brazilian underprivileged children. To create SIMPLEX-PB 2.0, we addressed all limitations of the old SIMPLEX-PB through multiple rounds of manual annotation. As a result, SIMPLEX-PB 2.0 features much more reliable and numerous candidate substitutions to complex words, as well as word complexity rankings produced by a group underprivileged children.

Automatic analysis of connected speech by natural language processing techniques is a promising direction for diagnosing cognitive impairments. However, some difficulties still remain: the time required for manual narrative transcription and the decision on how transcripts should be divided into sentences for successful application of parsers used in metrics, such as Idea Density, to analyze the transcripts. The main goal of this paper was to develop a generic segmentation system for narratives of neuropsychological language tests. We explored the performance of our previous single-dataset-trained sentence segmentation architecture in a richer scenario involving three new datasets used to diagnose cognitive impairments, comprising different stories and two types of stimulus presentation for eliciting narratives — visual and oral — via illustrated story-book and sequence of scenes, and by retelling. Also, we proposed and evaluated three modifications to our previous RCNN architecture: (i) the inclusion of a Linear Chain CRF; (ii) the inclusion of a self-attention mechanism; and (iii) the replacement of the LSTM recurrent layer by a Quasi-Recurrent Neural Network layer. Our study allowed us to develop two new models for segmenting impaired speech transcriptions, along with an ideal combination of datasets and specific groups of narratives to be used as the training set.

Effective textual communication depends on readers being proficient enough to comprehend texts, and texts being clear enough to be understood by the intended audience, in a reading task. When the meaning of textual information and instructions is not well conveyed, many losses and damages may occur. Among the solutions to alleviate this problem is the automatic evaluation of sentence readability, task which has been receiving a lot of attention due to its large applicability. However, a shortage of resources, such as corpora for training and evaluation, hinders the full development of this task. In this paper, we generate a nontrivial sentence corpus in Portuguese. We evaluate three scenarios for building it, taking advantage of a parallel corpus of simplification, in which each sentence triplet is aligned and has simplification operations annotated, being ideal for justifying possible mistakes of future methods. The best scenario of our corpus PorSimplesSent is composed of 4,888 pairs, which is bigger than a similar corpus for English; all the three versions of it are publicly available. We created four baselines for PorSimplesSent and made available a pairwise ranking method, using 17 linguistic and psycholinguistic features, which correctly identifies the ranking of sentence pairs with an accuracy of 74.2%.

Automated discourse analysis tools based on Natural Language Processing (NLP) aiming at the diagnosis of language-impairing dementias generally extract several textual metrics of narrative transcripts. However, the absence of sentence boundary segmentation in the transcripts prevents the direct application of NLP methods which rely on these marks in order to function properly, such as taggers and parsers. We present the first steps taken towards automatic neuropsychological evaluation based on narrative discourse analysis, presenting a new automatic sentence segmentation method for impaired speech. Our model uses recurrent convolutional neural networks with prosodic, Part of Speech (PoS) features, and word embeddings. It was evaluated intrinsically on impaired, spontaneous speech as well as normal, prepared speech and presents better results for healthy elderly (CTL) (F1 = 0.74) and Mild Cognitive Impairment (MCI) patients (F1 = 0.70) than the Conditional Random Fields method (F1 = 0.55 and 0.53, respectively) used in the same context of our study. The results suggest that our model is robust for impaired speech and can be used in automated discourse analysis tools to differentiate narratives produced by MCI and CTL.

Discriminating between Similar Languages (DSL) is a challenging task addressed at the VarDial Workshop series. We report on our participation in the DSL shared task with a two-stage system. In the first stage, character n-grams are used to separate language groups, then specialized classifiers distinguish similar language varieties. We have conducted experiments with three system configurations and submitted one run for each. Our main approach is a word-level convolutional neural network (CNN) that learns task-specific vectors with minimal text preprocessing. We also experiment with multi-layer perceptron (MLP) networks and another hybrid configuration. Our best run achieved an accuracy of 90.76%, ranking 8th among 11 participants and getting very close to the system that ranked first (less than 2 points). Even though the CNN model could not achieve the best results, it still makes a viable approach to discriminating between similar languages.

Mild Cognitive Impairment (MCI) is a mental disorder difficult to diagnose. Linguistic features, mainly from parsers, have been used to detect MCI, but this is not suitable for large-scale assessments. MCI disfluencies produce non-grammatical speech that requires manual or high precision automatic correction of transcripts. In this paper, we modeled transcripts into complex networks and enriched them with word embedding (CNE) to better represent short texts produced in neuropsychological assessments. The network measurements were applied with well-known classifiers to automatically identify MCI in transcripts, in a binary classification task. A comparison was made with the performance of traditional approaches using Bag of Words (BoW) and linguistic features for three datasets: DementiaBank in English, and Cinderella and Arizona-Battery in Portuguese. Overall, CNE provided higher accuracy than using only complex networks, while Support Vector Machine was superior to other classifiers. CNE provided the highest accuracies for DementiaBank and Cinderella, but BoW was more efficient for the Arizona-Battery dataset probably owing to its short narratives. The approach using linguistic features yielded higher accuracy if the transcriptions of the Cinderella dataset were manually revised. Taken together, the results indicate that complex networks enriched with embedding is promising for detecting MCI in large-scale assessments.

Portuguese is a less resourced language in what concerns foreign language learning. Aiming to inform a module of a system designed to support scientific written production of Spanish native speakers learning Portuguese, we developed an approach to automatically generate a lexicon of wrong words, reproducing language transfer errors made by such foreign learners. Each item of the artificially generated lexicon contains, besides the wrong word, the respective Spanish and Portuguese correct words. The wrong word is used to identify the interlanguage error and the correct Spanish and Portuguese forms are used to generate the suggestions. Keeping control of the correct word forms, we can provide correction or, at least, useful suggestions for the learners. We propose to combine two automatic procedures to obtain the error correction: i) a similarity measure and ii) a translation algorithm based on aligned parallel corpus. The similarity-based method achieved a precision of 52%, whereas the alignment-based method achieved a precision of 90%. In this paper we focus only on interlanguage errors involving suffixes that have different forms in both languages. The approach, however, is very promising to tackle other types of errors, such as gender errors.

Web 2.0 has allowed a never imagined communication boom. With the widespread use of computational and mobile devices, anyone, in practically any language, may post comments in the web. As such, formal language is not necessarily used. In fact, in these communicative situations, language is marked by the absence of more complex syntactic structures and the presence of internet slang, with missing diacritics, repetitions of vowels, and the use of chat-speak style abbreviations, emoticons and colloquial expressions. Such language use poses severe new challenges for Natural Language Processing (NLP) tools and applications, which, so far, have focused on well-written texts. In this work, we report the construction of a large web corpus of product reviews in Brazilian Portuguese and the analysis of its lexical phenomena, which support the development of a lexical normalization tool for, in future work, subsidizing the use of standard NLP products for web opinion mining and summarization purposes.

This paper reports the annotation of a Brazilian Portuguese Treebank with semantic role labels following Propbank guidelines. A different language and a different parser output impact the task and require some decisions on how to annotate the corpus. Therefore, a new annotation guide ― called Propbank-Br - has been generated to deal with specific language phenomena and parser problems. In this phase of the project, the corpus was annotated by a unique linguist. The annotation task reported here is inserted in a larger projet for the Brazilian Portuguese language. This project aims to build Brazilian verbs frames files and a broader and distributed annotation of semantic role labels in Brazilian Portuguese, allowing inter-annotator agreement measures. The corpus, available in web, is already being used to build a semantic tagger for Portuguese language.

The relevance of automatically identifying rhetorical moves in scientific texts has been widely acknowledged in the literature. This study focuses on abstracts of standard research papers written in English and aims to tackle a fundamental limitation of current machine-learning classifiers: they are mono-labeled, that is, a sentence can only be assigned one single label. However, such approach does not adequately reflect actual language use since a move can be realized by a clause, a sentence, or even several sentences. Here, we present MAZEA (Multi-label Argumentative Zoning for English Abstracts), a multi-label classifier which automatically identifies rhetorical moves in abstracts but allows for a given sentence to be assigned as many labels as appropriate. We have resorted to various other NLP tools and used two large training corpora: (i) one corpus consists of 645 abstracts from physical sciences and engineering (PE) and (ii) the other corpus is made up of 690 from life and health sciences (LH). This paper presents our preliminary results and also discusses the various challenges involved in multi-label tagging and works towards satisfactory solutions. In addition, we also make our two training corpora publicly available so that they may serve as benchmark for this new task.

This work reports the evaluation and selection of annotation tools to assign wh-question labels to verbal arguments in a sentence. Wh-question assignment discussed herein is a kind of semantic annotation which involves two tasks: making delimitation of verbs and arguments, and linking verbs to its arguments by question labels. As it is a new type of semantic annotation, there is no report about requirements an annotation tool should have to face it. For this reason, we decided to select the most appropriated tool in two phases. In the first phase, we executed the task with an annotation tool we have used before in another task. Such phase helped us to test the task and enabled us to know which features were or not desirable in an annotation tool for our purpose. In the second phase, guided by such requirements, we evaluated several tools and selected a tool for the real task. After corpus annotation conclusion, we report some of the annotation results and some comments on the improvements there should be made in an annotation tool to better support such kind of annotation task.

In this paper we discuss the five requirements for building large publicly available corpora which geared the construction of the Lácio-Web corpora and their environments: 1) a comprehensive text typology; 2) text copyright clearance, compilation and annotation scheme; 3) a friendly and didactic interface; 4) the need to serve as support for several types of research; 5) the need to offer an array of associated tools. Also, we present the features that make Lácio-Web corpora interesting and novel as well as the limitations of this project, such as corpora size and balance, and the non-inclusion of spoken texts in the project’s reference corpus.