In recent decades, statistical approaches have significantly advanced the development of machine translation systems. However, the applicability of these methods directly depends on the availability of very large quantities of parallel data. Recent works have demonstrated that a comparable corpus can compensate for the shortage of parallel corpora. In this paper, we propose an alternative to comparable corpora containing text documents as resources for extracting parallel data: a multimodal comparable corpus with audio documents in source language and text document in target language, built from Euronews and TED web sites. The audio is transcribed by an automatic speech recognition system, and translated with a baseline statistical machine translation system. We then use information retrieval in a large text corpus in the target language in order to extract parallel sentences/phrases. We evaluate the quality of the extracted data on an English to French translation task and show significant improvements over a state-of-the-art baseline.

There has been considerable work recently in the natural language community and elsewhere on Responsible AI. Much of this work focuses on fairness and biases (henceforth Risks 1.0), following the 2016 best seller: Weapons of Math Destruction. Two books published in 2022, The Chaos Machine and Like, Comment, Subscribe, raise additional risks to public health/safety/security such as genocide, insurrection, polarized politics, vaccinations (henceforth, Risks 2.0). These books suggest that the use of machine learning to maximize engagement in social media has created a Frankenstein Monster that is exploiting human weaknesses with persuasive technology, the illusory truth effect, Pavlovian conditioning, and Skinner’s intermittent variable reinforcement. Just as we cannot expect tobacco companies to sell fewer cigarettes and prioritize public health ahead of profits, so too, it may be asking too much of companies (and countries) to stop trafficking in misinformation given that it is so effective and so insanely profitable (at least in the short term). Eventually, we believe the current chaos will end, like the lawlessness in Wild West, because chaos is bad for business. As computer scientists, this paper will summarize criticisms from other fields and focus on implications for computer science; we will not attempt to contribute to those other fields. There is quite a bit of work in computer science on these risks, especially on Risks 1.0 (bias and fairness), but more work is needed, especially on Risks 2.0 (addictive, dangerous, and deadly).

Learning idiomatic expressions is seen as one of the most challenging stages in second-language learning because of their unpredictable meaning. A similar situation holds for their identification within natural language processing applications such as machine translation and parsing. The lack of high-quality usage samples exacerbates this challenge not only for humans but also for artificial intelligence systems. This article introduces a gamified crowdsourcing approach for collecting language learning materials for idiomatic expressions; a messaging bot is designed as an asynchronous multiplayer game for native speakers who compete with each other while providing idiomatic and nonidiomatic usage examples and rating other players’ entries. As opposed to classical crowd-processing annotation efforts in the field, for the first time in the literature, a crowd-creating & crowd-rating approach is implemented and tested for idiom corpora construction. The approach is language-independent and evaluated on two languages in comparison to traditional data preparation techniques in the field. The reaction of the crowd is monitored under different motivational means (namely, gamification affordances and monetary rewards). The results reveal that the proposed approach is powerful in collecting the targeted materials, and although being an explicit crowdsourcing approach, it is found entertaining and useful by the crowd. The approach has been shown to have the potential to speed up the construction of idiom corpora for different natural languages to be used as second-language learning material, training data for supervised idiom identification systems, or samples for lexicographic studies.

Automatic text summarisation is a topic that has been receiving attention from the research community from the early days of computational linguistics, but it really took off around 25 years ago. This article presents the main developments from the last 25 years. It starts by defining what a summary is and how its definition changed over time as a result of the interest in processing new types of documents. The article continues with a brief history of the field and highlights the main challenges posed by the evaluation of summaries. The article finishes with some thoughts about the future of the field.

This article provides a global overview of the main aspects of current practice in the design, implementation and evaluation of speech recognition components for Spoken Language Dialog Systems (SLDSs), and presents the results of the DISC European project related to speech recognition. DISC and its successor DISC-2 are efforts towards the definition of best practice guidelines for SLDS development and evaluation. SLDSs aim at using natural spoken input for performing an information processing task such as automated standards, call routing or travel planning and reservations. The main functionality of an SLDS are speech recognition, natural language understanding, dialog management, database access and interpretation, response generation and speech synthesis. Speech recognition, which transforms the acoustic signal into a string of words, is a key technology in any SLDS.

This paper starts by introducing a class of future document authoring systems that will allow authors to specify the content and form of a text+pictures document at a high level of abstraction, while leaving responsibility for linguistic and graphical details to the system. Next, we describe two working prototypes that implement parts of this functionality, based on semantic modeling of the pictures and the text of the document; one of these two, the ILLUSTRATE prototype, is a multimedia extension of previous text authoring systems in the What You See Is What You Meant (WYSIWYM) tradition. The paper concludes with an exploration of the ways in which Multimedia WYSIWYM can be further enhanced, allowing it to approximate the ‘ideal’ systems that were sketched earlier in the paper. Applications of Multimedia WYSIWYM to general-purpose picture retrieval (in the context of the Semantic Web, for example) are also discussed.

An open problem in natural language processing is word sense disambiguation (WSD). A word may have several meanings, but WSD is the task of selecting the correct sense of a polysemous word based on its context. Proposed solutions are based on supervised and unsupervised learning methods. The majority of researchers in the area focused on choosing proper size of ‘n’ in n-gram that is used for WSD problem. In this research, the concept has been taken to a new level by using variable ‘n’ and variable size window. The concept is based on the iterative patterns extracted from the text. We show that this type of sequential pattern is more effective than many other solutions for WSD. Using regular data mining algorithms on the extracted features, we significantly outperformed most monolingual WSD solutions. The state-of-the-art results were obtained using external knowledge like various translations of the same sentence. Our method improved the accuracy of the multilingual system more than 4 percent, although we were using monolingual features.

An algorithmic approach to the semantic interpretation of deverbal nominalizations found in encyclopedic texts, such as support, publication and control, is described. Interpreting these nominalizations is crucial because they are quite common in encyclopedic texts, hence a great deal of information is represented within them. Interpretation involves three tasks: deciding whether the nominalization is being used in a verbal or non-verbal sense; disambiguating the nominalized verb when a verbal sense is used; and determining the fillers of the thematic roles of the verbal concept or predicate of the nominalization. A verbal sense can be recognized by the presence of modifiers that represent the arguments of the verbal concept. It is these same modifiers which provide the semantic clues to disambiguate the nominalized verb. In the absence of explicit modifiers, heuristics are used to discriminate between verbal and non-verbal senses. A correspondence between verbs and their nominalizations is exploited so that only a small amount of additional knowledge is needed to handle the nominal form. These methods are tested in the domain of encyclopedic texts and the results are shown.

Approaches to building temporal information extraction systems typically rely on large, manually annotated corpora. Thus, porting these systems to new languages requires acquiring large corpora of manually annotated documents in the new languages. Acquiring such corpora is difficult owing to the complexity of temporal information extraction annotation. One strategy for addressing this difficulty is to reduce or eliminate the need for manually annotated corpora through annotation projection. This technique utilizes a temporal information extraction system for a source language (typically English) to automatically annotate the source language side of a parallel corpus. It then uses automatically generated word alignments to project the annotations, thereby creating noisily annotated target language training data. We developed an annotation projection technique for producing target language temporal information extraction systems. We carried out an English (source) to French (target) case study wherein we compared a French temporal information extraction system built using annotation projection with one built using a manually annotated French corpus. While annotation projection has been applied to building other kinds of Natural Language Processing tools (e.g., Named Entity Recognizers), to our knowledge, this is the first paper examining annotation projection as applied to temporal information extraction where no manual corrections of the target language annotations were made. We found that, even using manually annotated data to build a temporal information extraction system, F-scores were relatively low (<0.35), which suggests that the problem is challenging even with manually annotated data. Our annotation projection approach performed well (relative to the system built from manually annotated data) on some aspects of temporal information extraction (e.g., event–document creation time temporal relation prediction), but it performed poorly on the other kinds of temporal relation prediction (e.g., event–event and event–time).

This article presents a new method to automatically simplify English sentences. The approach is designed to reduce the number of compound clauses and nominally bound relative clauses in input sentences. The article provides an overview of a corpus annotated with information about various explicit signs of syntactic complexity and describes the two major components of a sentence simplification method that works by exploiting information on the signs occurring in the sentences of a text. The first component is a sign tagger which automatically classifies signs in accordance with the annotation scheme used to annotate the corpus. The second component is an iterative rule-based sentence transformation tool. Exploiting the sign tagger in conjunction with other NLP components, the sentence transformation tool automatically rewrites long sentences containing compound clauses and nominally bound relative clauses as sequences of shorter single-clause sentences. Evaluation of the different components reveals acceptable performance in rewriting sentences containing compound clauses but less accuracy when rewriting sentences containing nominally bound relative clauses. A detailed error analysis revealed that the major sources of error include inaccurate sign tagging, the relatively limited coverage of the rules used to rewrite sentences, and an inability to discriminate between various subtypes of clause coordination. Despite this, the system performed well in comparison with two baselines. This finding was reinforced by automatic estimations of the readability of system output and by surveys of readers’ opinions about the accuracy, accessibility, and meaning of this output.

This article reports on designing and implementing a multiclass sentiment classification approach to handle the imbalanced class distribution of Arabic documents. The proposed approach, sentiment classification of Arabic documents (SCArD), combines the advantages of a clustering-based undersampling (CBUS) method and an ensemble learning model to aid machine learning (ML) classifiers in building accurate models against highly imbalanced datasets. The CBUS method applies two standard clustering algorithms: K-means and expectation–maximization, to balance the ratio between the major and the minor classes by decreasing the number of the major class instances and maintaining the number of the minor class instances at the cluster level. The merits of the proposed approach are that it does not remove the majority class instances from the dataset nor injects the dataset with artificial minority class instances. The resulting balanced datasets are used to train two ML classifiers, random forest and updateable Naïve Bayes, to develop prediction data models. The best prediction data models are selected based on F1-score rates. We applied two techniques to test SCArD and generate new predictions from the imbalanced test dataset. The first technique uses the best prediction data models. The second technique uses the majority voting ensemble learning model, which combines the best prediction data models to generate the final predictions. The experimental results showed that SCArD is promising and outperformed the other comparative classification models based on the F1-score rates.

We consider a dependency-parsed text corpus as an instance of a labeled directed graph, where nodes represent words and weighted directed edges represent the syntactic relations between them. We show that graph walks, combined with existing techniques of supervised learning that model local and global information about the graph walk process, can be used to derive a task-specific word similarity measure in this graph. We also propose and evaluate a new learning method in this framework, a path-constrained graph walk variant, in which the walk process is guided by high-level knowledge about meaningful edge sequences (paths) in the graph. Empirical evaluation on the tasks of named entity coordinate term extraction and general word synonym extraction show that this framework is preferable to, or competitive with, vector-based models when learning is applied, and using small to moderate size text corpora.

Automatic term extraction has become a priority area of research within corpus processing. Despite the extensive literature in this field, there are still some outstanding issues that should be dealt with during the construction of term extractors, particularly those oriented to support research in terminology and terminography. In this regard, this article describes the design and development of DEXTER, an online workbench for the extraction of simple and complex terms from domain-specific corpora in English, French, Italian and Spanish. In this framework, three issues contribute to placing the most important terms in the foreground. First, unlike the elaborate morphosyntactic patterns proposed by most previous research, shallow lexical filters have been constructed to discard term candidates. Second, a large number of common stopwords are automatically detected by means of a method that relies on the IATE database together with the frequency distribution of the domain-specific corpus and a general corpus. Third, the term-ranking metric, which is grounded on the notions of salience, relevance and cohesion, is guided by the IATE database to display an adequate distribution of terms.

The user-generated Web content has been intensively analyzed in Information Extraction and Natural Language Processing research. Web-posted reviews of consumer goods are studied to find customer opinions about the products. We hypothesize that nonemotionally charged descriptions can be applied to predict those opinions. The descriptions may include indicators of product size (tall), commonplace (some), frequency of happening (often), and reviewer certainty (maybe). We first construct patterns of how the descriptions are used in consumer-written texts and then represent individual reviews through these patterns. We propose a semantic hierarchy that organizes individual words into opinion types. We run machine learning algorithms on five data sets of user-written product reviews: four are used in classification experiments, another one for regression and classification. The obtained results support the use of non-emotional descriptions in opinion learning.

The paper reviews the state of the art of natural language engineering (NLE) around 1995, when this journal first appeared, and makes a critical comparison with the current state of the art in 2018, as we prepare the 25th Volume. Specifically the then state of the art in parsing, information extraction, chatbots, and dialogue systems, speech processing and machine translation are briefly reviewed. The emergence in the 1980s and 1990s of machine learning (ML) and statistical methods (SM) is noted. Important trends and areas of progress in the subsequent years are identified. In particular, the move to the use of n-grams or skip grams and/or chunking with part of speech tagging and away from whole sentence parsing is noted, as is the increasing dominance of SM and ML. Some outstanding issues which merit further research are briefly pointed out, including metaphor processing and the ethical implications of NLE.

It is amazing that I have written for more than a year on Emerging Trends without mentioning China’s investments in Artificial Intelligence (AI). Now that I have moved to Baidu, I would like to take this opportunity to share some of my personal observations with what’s happening in China over the past 25 years. The top universities in China have always been very good, but they are better today than they were 25 years ago, and they are on a trajectory to become the biggest and the best in the world. China is investing big time in what we do, both in the private sector and the public sector. Kai-Fu Lee is bullish on his investments in AI and China. There is a bold government plan for AI with specific milestones for parity with the West in 2020, major breakthroughs by 2025 and the envy of the world by 2030.

Sentence embeddings are the suitable input vectors for the neural learning of a number of inferences about content and meaning. Similarity estimation, classification, emotional characterization of sentences as well as pragmatic tasks, such as question answering or dialogue, have largely demonstrated the effectiveness of vector embeddings to model semantics. Unfortunately, most of the above decisions are epistemologically opaque as for the limited interpretability of the acquired neural models based on the involved embeddings. We think that any effective approach to meaning representation should be at least epistemologically coherent. In this paper, we concentrate on the readability of neural models, as a core property of any embedding technique consistent and effective in representing sentence meaning. In this perspective, this paper discusses a novel embedding technique (the Nyström methodology) that corresponds to the reconstruction of a sentence in a kernel space, inspired by rich semantic similarity metrics (a semantic kernel) rather than by a language model. In addition to being based on a kernel that captures grammatical and lexical semantic information, the proposed embedding can be used as the input vector of an effective neural learning architecture, called Kernel-based deep architectures (KDA). Finally, it also characterizes by design the KDA explanatory capability, as the proposed embedding is derived from examples that are both human readable and labeled. This property is obtained by the integration of KDAs with an explanation methodology, called layer-wise relevance propagation (LRP), already proposed in image processing. The Nyström embeddings support here the automatic compilation of argumentations in favor or against a KDA inference, in form of an explanation: each decision can in fact be linked through LRP back to the real examples, that is, the landmarks linguistically related to the input instance. The KDA network output is explained via the analogy with the activated landmarks. Quantitative evaluation of the explanations shows that richer explanations based on semantic and syntagmatic structures characterize convincing arguments, as they effectively help the user in assessing whether or not to trust the machine decisions in different tasks, for example, Question Classification or Semantic Role Labeling. This confirms the epistemological benefit that Nyström embeddings may bring, as linguistically rich and meaningful representations for a variety of inference tasks.

We present a lexical platform that has been developed for the Spanish language. It achieves portability between different computer systems and efficiency, in terms of speed and lexical coverage. A model for the full treatment of Spanish inflectional morphology for verbs, nouns and adjectives is presented. This model permits word formation based solely on morpheme concatenation, driven by a feature-based unification grammar. The run-time lexicon is a collection of allomorphs for both stems and endings. Although not tested, it should be suitable also for other Romance and highly inflected languages. A formalism is also described for encoding a lemma-based lexical source, well suited for expressing linguistic generalizations: inheritance classes, lemma encoding, morpho-graphemic allomorphy rules and limited type-checking. From this source base, we can automatically generate an allomorph indexed dictionary adequate for efficient retrieval and processing. A set of software tools has been implemented around this formalism: lexical base augmenting aids, lexical compilers to build run-time dictionaries and access libraries for them, feature manipulation libraries, unification and pseudo-unification modules, morphological processors, a parsing system, etc. Software interfaces among the different modules and tools are cleanly defined to ease software integration and tool combination in a flexible way. Directions for accessing our e-mail and web demonstration prototypes are also provided. Some figures are given, showing the lexical coverage of our platform compared to some popular spelling checkers.

We present research aiming to build tools for the normalization of User-Generated Content (UGC). We argue that processing this type of text requires the revisiting of the initial steps of Natural Language Processing, since UGC (micro-blog, blog, and, generally, Web 2.0 user-generated texts) presents a number of nonstandard communicative and linguistic characteristics – often closer to oral and colloquial language than to edited text. We present a corpus of UGC text in Spanish from three different sources: Twitter, consumer reviews, and blogs, and describe its main characteristics. We motivate the need for UGC text normalization by analyzing the problems found when processing this type of text through a conventional language processing pipeline, particularly in the tasks of lemmatization and morphosyntactic tagging. Our aim with this paper is to seize the power of already existing spell and grammar correction engines and endow them with automatic normalization capabilities in order to pave the way for the application of standard Natural Language Processing tools to typical UGC text. Particularly, we propose a strategy for automatically normalizing UGC by adding a module on top of a pre-existing spell-checker that selects the most plausible correction from an unranked list of candidates provided by the spell-checker. To build this selector module we train four language models, each one containing a different type of linguistic information in a trade-off with its generalization capabilities. Our experiments show that the models trained on truecase and lowercase word forms are more discriminative than the others at selecting the best candidate. We have also experimented with a parametrized combination of the models by both optimizing directly on the selection task and doing a linear interpolation of the models. The resulting parametrized combinations obtain results close to the best performing model but do not improve on those results, as measured on the test set. The precision of the selector module in ranking number one the expected correction proposal on the test corpora reaches 82.5% for Twitter text (baseline 57%) and 88% for non-Twitter text (baseline 64%).

In source reordering the order of the source words is permuted to minimize word order differences with the target sentence and then fed to a translation model. Earlier work highlights the benefits of resolving long-distance reorderings as a pre-processing step to standard phrase-based models. However, the potential performance improvement of source reordering and its impact on the components of the subsequent translation model remain unexplored. In this paper we study both aspects of source reordering. We set up idealized source reordering (oracle) models with/without syntax and present our own syntax-driven model of source reordering. The latter is a statistical model of inversion transduction grammar (ITG)-like tree transductions manipulating a syntactic parse and working with novel conditional reordering parameters. Having set up the models, we report translation experiments showing significant improvement on three language pairs, and contribute an extensive analysis of the impact of source reordering (both oracle and model) on the translation model regarding the quality of its input, phrase-table, and output. Our experiments show that oracle source reordering has untapped potential in improving translation system output. Besides solving difficult reorderings, we find that source reordering creates more monotone parallel training data at the back-end, leading to significantly larger phrase tables with higher coverage of phrase types in unseen data. Unfortunately, this nice property does not carry over to tree-constrained source reordering. Our analysis shows that, from the string-level perspective, tree-constrained reordering might selectively permute word order, leading to larger phrase tables but without increase in phrase coverage in unseen data.