José Camacho-Collados

• MS
• Lecturer at Cardiff University

BACKGROUND Free-text clinical data that is unstructured and narrative in nature can provide a rich source of patient information. However, the information contained within routinely collected health data is typically captured as free-text, and extracting research quality clinical phenotypes from these data remains a challenge. Manually reviewing an...

Extracting metaphors and analogies from free text requires high-level reasoning abilities such as abstraction and language understanding. Our study focuses on the extraction of the concepts that form metaphoric analogies in literary texts. To this end, we construct a novel dataset in this domain with the help of domain experts. We compare the out-o...

The detection of sensitive content in large datasets is crucial for ensuring that shared and analysed data is free from harmful material. However, current moderation tools, such as external APIs, suffer from limitations in customisation, accuracy across diverse sensitive categories, and privacy concerns. Additionally, existing datasets and open-sou...

The metaphor studies community has developed numerous valuable labelled corpora in various languages over the years. Many of these resources are not only unknown to the NLP community, but are also often not easily shared among the researchers. Both in human sciences and in NLP, researchers could benefit from a centralised database of labelled resou...

Background: Mental health disorders are currently the main contributor to poor quality of life and years lived with disability. Symptoms common to many mental health disorders lead to impairments or changes in the use of language, which are observable in the routine use of social media. Detection of these linguistic cues has been explored throughou...

Social media offers the potential to provide detection of outbreaks or public health incidents faster than traditional reporting mechanisms. In this paper, we developed and tested a pipeline to produce alerts of influenza-like illness (ILI) using Twitter data. Data was collected from the Twitter API, querying keywords referring to ILI symptoms and...

In the dynamic realm of social media, diverse topics are discussed daily, transcending linguistic boundaries. However, the complexities of understanding and categorising this content across various languages remain an important challenge with traditional techniques like topic modelling often struggling to accommodate this multilingual diversity. In...

Readability-controlled text simplification (RCTS) rewrites texts to lower readability levels while preserving their meaning. RCTS models often depend on parallel corpora with readability annotations on both source and target sides. Such datasets are scarce and difficult to curate, especially at the sentence level. To reduce reliance on parallel dat...

Collecting and analysing data after an earthquake is essential to determine its impact. In 2014 The European Mediterranean Seismological Centre (EMSC) launched a multichannel rapid information system comprising websites, a Twitter quakebot, and a smartphone app for global earthquake eyewitnesses: the LastQuake system. This system collects intensity...

Social biases such as gender or racial biases have been reported in language models (LMs), including Masked Language Models (MLMs). Given that MLMs are continuously trained with increasing amounts of additional data collected over time, an important yet unanswered question is how the social biases encoded with MLMs vary over time. In particular, th...

Large language models (LLMs) often lack culture-specific knowledge of daily life, especially across diverse regions and non-English languages. Existing benchmarks for evaluating LLMs' cultural sensitivities are limited to a single language or collected from online sources such as Wikipedia, which do not reflect the mundane everyday lifestyles of di...

Trigger points are a concept introduced by Mau, Lux, and Westheuser (2023) to study qualitative focus group interviews and understand polarisation in Germany. When people communicate, trigger points represent moments when individuals feel that their understanding of what is fair, normal, or appropriate in society is questioned. In the original stud...

The annotation of ambiguous or subjective NLP tasks is usually addressed by various annotators. In most datasets, these annotations are aggregated into a single ground truth. However, this omits divergent opinions of annotators, hence missing individual perspectives. We propose FLEAD (Federated Learning for Exploiting Annotators’ Disagreements), a...

In machine learning, temporal shifts occur when there are differences between training and test splits in terms of time. For streaming data such as news or social media, models are commonly trained on a fixed corpus from a certain period of time, and they can become obsolete due to the dynamism and evolving nature of online content. This paper focu...

BACKGROUND Mental health disorders are currently the main contributor to poor quality of life and years lived with disability. Symptoms common to many mental health disorders lead to impairments or changes in the use of language, which are observable in the routine use of social media. Detection of these linguistic cues has been explored throughout...

Precision medicine has the ambition to improve treatment response and clinical outcomes through patient stratification, and holds great potential in mental disorders. However, several important factors are needed to transform current practice into a “precision psychiatry” framework. Most important are (1) the generation of accessible large real-wor...

Recovery is a complex, multidimensional long-term process of restoration of living conditions after a disaster. Following an earthquake, there is substantial demand for information. However, data collection usually occurs only during the emergency phase. The absence of data afterwards has resulted in limited knowledge of earthquakes' long-term imp...

Recovery is a complex multidimensional long-term process of restoration of living conditions after a disaster. Following an earthquake, there is substantial demand for information. However, data collection usually occurs only during the emergency phase. The absence of data afterwards has resulted in limited knowledge of earthquakes' long-term impac...

We describe the first edition of the LongEval CLEF 2023 shared task. This lab evaluates the temporal persistence of Information Retrieval (IR) systems and Text Classifiers. Task 1 requires IR systems to run on corpora acquired at several timestamps, and evaluates the drop in system quality (NDCG) along these timestamps. Task 2 tackles binary sentim...

We describe the first edition of the LongEval CLEF 2023 shared task. This lab evaluates the temporal persistence of Information Retrieval (IR) systems and Text Classifiers. Task 1 requires IR systems to run on corpora acquired at several timestamps, and evaluates the drop in system quality (NDCG) along these timestamps. Task 2 tackles binary sentim...

This paper introduces a large collection of time series data derived from Twitter, postprocessed using word embedding techniques, as well as specialized fine-tuned language models. This data comprises the past five years and captures changes in n-gram frequency, similarity, sentiment and topic distribution. The interface built on top of this data e...

The automatic detection of hate speech online is an active research area in NLP. Most of the studies to date are based on social media datasets that contribute to the creation of hate speech detection models trained on them. However, data creation processes contain their own biases, and models inherently learn from these dataset-specific biases. In...

Hate speech is one of the major concerns of Europe. Different studies, mainly in English language, have been carried out to analyze hate speech, many of them from a theoretical perspective. Here, it is presented an observational study about hate speech poured on Twitter in Spanish regarding to five social important events: Women´s Day, Internationa...

Generating questions along with associated answers from a text has applications in several domains, such as creating reading comprehension tests for students, or improving document search by providing auxiliary questions and answers based on the query. Training models for question and answer generation (QAG) is not straightforward due to the expect...

Question and answer generation (QAG) consists of generating a set of question-answer pairs given a context (e.g. a paragraph). This task has a variety of applications, such as data augmentation for question answering (QA) models, information retrieval and education. In this paper, we establish baselines with three different QAG methodologies that l...

Multilingual language model (LM) have become a powerful tool in NLP especially for non-English languages. Nevertheless, model parameters of multilingual LMs remain large due to the larger embedding matrix of the vocabulary covering tokens in different languages. On the contrary, monolingual LMs can be trained in a target language with the language-...

Relations such as "is influenced by", "is known for" or "is a competitor of" are inherently graded: we can rank entity pairs based on how well they satisfy these relations, but it is hard to draw a line between those pairs that satisfy them and those that do not. Such graded relations play a central role in many applications, yet they are typically...

Background: Major depressive disorder is a common mental disorder affecting 5% of adults worldwide. Early contact with health care services is critical for achieving accurate diagnosis and improving patient outcomes. Key symptoms of major depressive disorder (depression hereafter) such as cognitive distortions are observed in verbal communication,...

In this paper, we describe the plans for the first LongEval CLEF 2023 shared task dedicated to evaluating the temporal persistence of Information Retrieval (IR) systems and Text Classifiers. The task is motivated by recent research showing that the performance of these models drops as the test data becomes more distant, with respect to time, from t...

Powerful generative models have led to recent progress in question generation (QG). However, it is difficult to measure advances in QG research since there are no standardized resources that allow a uniform comparison among approaches. In this paper, we introduce QG-Bench, a multilingual and multidomain benchmark for QG that unifies existing questi...

Recent progress in language model pre-training has led to important improvements in Named Entity Recognition (NER). Nonetheless, this progress has been mainly tested in well-formatted documents such as news, Wikipedia, or scientific articles. In social media the landscape is different, in which it adds another layer of complexity due to its noisy a...

Social media platforms host discussions about a wide variety of topics that arise everyday. Making sense of all the content and organising it into categories is an arduous task. A common way to deal with this issue is relying on topic modeling, but topics discovered using this technique are difficult to interpret and can differ from corpus to corpu...

Language evolves over time, and word meaning changes accordingly. This is especially true in social media, since its dynamic nature leads to faster semantic shifts, making it challenging for NLP models to deal with new content and trends. However, the number of datasets and models that specifically address the dynamic nature of these social platfor...

Language model (LM) pretraining has led to consistent improvements in many NLP downstream tasks, including named entity recognition (NER). In this paper, we present T-NER (Transformer-based Named Entity Recognition), a Python library for NER LM finetuning. In addition to its practical utility, T-NER facilitates the study and investigation of the cr...

BACKGROUND Major Depressive Disorder (MDD) is a common mental disorder that affects 5% of adults worldwide. Early contact with healthcare services is critical in achieving an accurate diagnosis and improving patient outcomes. Key symptoms of MDD (depression hereafter) such as cognitive distortions are observed in verbal communication, which can man...

In this paper we present TweetNLP, an integrated platform for Natural Language Processing (NLP) in social media. TweetNLP supports a diverse set of NLP tasks, including generic focus areas such as sentiment analysis and named entity recognition, as well as social media-specific tasks such as emoji prediction and offensive language identification. T...

The increase in performance in NLP due to the prevalence of distributional models and deep learning has brought with it a reciprocal decrease in interpretability. This has spurred a focus on what neural networks learn about natural language with less of a focus on how. Some work has focused on the data used to develop data-driven models, but typica...

Despite its importance, the time variable has been largely neglected in the NLP and language model literature. In this paper, we present TimeLMs, a set of language models specialized on diachronic Twitter data. We show that a continual learning strategy contributes to enhancing Twitter-based language models' capacity to deal with future and out-of-...

Social media has become extremely influential when it comes to policy making in modern societies especially in the western world (e.g., 48% of Europeans use social media every day or almost every day). Platforms such as Twitter allow users to follow politicians, thus making citizens more involved in political discussion. In the same vein, politicia...

Distributional semantics based on neural approaches is a cornerstone of Natural Language Processing, with surprising connections to human meaning representation as well. Recent Transformer-based Language Models have proven capable of producing contextual word representations that reliably convey sense-specific information, simply as a product of se...

Data augmentation techniques are widely used for enhancing the performance of machine learning models by tackling class imbalance issues and data sparsity. State-of-the-art generative language models have been shown to provide significant gains across different NLP tasks. However, their applicability to data augmentation for text classification tas...

Word embeddings have become a standard resource in the toolset of any Natural Language Processing practitioner. While monolingual word embeddings encode information about words in the context of a particular language, cross-lingual embeddings define a multilingual space where word embeddings from two or more languages are integrated together. Curre...

Pre-trained language models have been found to capture a surprisingly rich amount of lexical knowledge, ranging from commonsense properties of everyday concepts to detailed factual knowledge about named entities. Among others, this makes it possible to distill high-quality word vectors from pre-trained language models. However, it is currently uncl...

This paper demonstrates a two-stage method for deriving insights from social media data relating to disinformation by applying a combination of geospatial classification and embedding-based language modelling across multiple languages. In particular, the analysis in centered on Twitter and disinformation for three European languages: English, Frenc...

While the success of pre-trained language models has largely eliminated the need for high-quality static word vectors in many NLP applications, static word vectors continue to play an important role in tasks where word meaning needs to be modelled in the absence of linguistic context. In this paper, we explore how the contextualised embeddings pred...

Transformer-based language models have taken many fields in NLP by storm. BERT and its derivatives dominate most of the existing evaluation benchmarks, including those for Word Sense Disambiguation (WSD), thanks to their ability in capturing context-sensitive semantic nuances. However, there is still little knowledge about their capabilities and po...

Distributional semantics based on neural approaches is a cornerstone of Natural Language Processing, with surprising connections to human meaning representation as well. Recent Transformer-based Language Models have proven capable of producing contextual word representations that reliably convey sense-specific information, simply as a product of se...

Analogies play a central role in human commonsense reasoning. The ability to recognize analogies such as eye is to seeing what ear is to hearing, sometimes referred to as analogical proportions, shape how we structure knowledge and understand language. Surprisingly, however, the task of identifying such analogies has not yet received much attention...

Language models are ubiquitous in current NLP, and their multilingual capacity has recently attracted considerable attention. However, current analyses have almost exclusively focused on (multilingual variants of) standard benchmarks, and have relied on clean pre-training and task-specific corpora as multilingual signals. In this paper, we introduc...

Term weighting schemes are widely used in Natural Language Processing and Information Retrieval. In particular, term weighting is the basis for keyword extraction. However, there are relatively few evaluation studies that shed light about the strengths and shortcomings of each weighting scheme. In fact, in most cases researchers and practitioners r...

Most current Machine Learning models are data-driven: they learn from the data to which they are exposed. Therefore, they inevitably encode all the implicit stereotyped biases, such as gender, racial, or ideological biases, present in the data, unless specific measures are undertaken to prevent this. This raises concerns over the use of these techn...

In this chapter, we introduce a type representation aimed at modeling unambiguous lexical items.1 These representations emerged in order to address one of the main limitations of word-level representation techniques, meaning conflation.

Graphs are ubiquitous data structures. They are often the preferred choice for representing various type of data, including social networks, word co-occurrence and semantic networks, citation networks, telecommunication networks, molecular graph structures, and biological networks. Therefore, analyzing them can play a central role in various real-w...

In the first part of the book, we focused on some of the smallest units in language, mostly those at the word-level. However, in most applications dealing with natural language, understanding longer units of meaning such as sentences1 and documents is crucial. While this chapter is not covered as exhaustively as the other chapters, we still provide...

Section 1.2 briefly discussed the Vector Space Model (VSM). We saw in that section how objects can be represented using continuous vectors in a multidimensional space and how distances in this space can denote the similarities between objects. However, we did not discuss how these spaces are constructed. In other words, the following question remai...

This chapter provides an introduction to contextualized word embeddings which can be considered the new generation of word (and sense) embeddings. The distinguishing factor here is the sensitivity of a word’s representation to the context: a target word’s embedding can change depending on the context in which it appears. These dynamic embeddings al...

Depression and anxiety are psychiatric disorders that are observed in many areas of everyday life. For example, these disorders manifest themselves somewhat frequently in texts written by non-diagnosed users in social media. However, detecting users with these conditions is not a straightforward task as they may not explicitly talk about their ment...

While the success of pre-trained language models has largely eliminated the need for high-quality static word vectors in many NLP applications, static word vectors continue to play an important role in tasks where word meaning needs to be modelled in the absence of linguistic context. In this paper, we explore how the contextualised embeddings pred...

Depression and anxiety are psychiatric disorders that are observed in many areas of everyday life. For example, these disorders manifest themselves somewhat frequently in texts written by nondiagnosed users in social media. However, detecting users with these conditions is not a straightforward task as they may not explicitly talk about their menta...

The task of text and sentence classification is associated with the need for large amounts of labelled training data. The acquisition of high volumes of labelled datasets can be expensive or unfeasible, especially for highly-specialised domains for which documents are hard to obtain. Research on the application of supervised classification based on...

The experimental landscape in natural language processing for social media is too fragmented. Each year, new shared tasks and datasets are proposed, ranging from classics like sentiment analysis to irony detection or emoji prediction. Therefore, it is unclear what the current state of the art is, as there is no standardized evaluation protocol, nei...

The ability to correctly model distinct meanings of a word is crucial for the effectiveness of semantic representation techniques. However, most existing evaluation benchmarks for assessing this criterion are tied to sense inventories (usually WordNet), restricting their usage to a small subset of knowledge-based representation techniques. The Word...

Transformer-based language models have taken many fields in NLP by storm. BERT and its derivatives dominate most of the existing evaluation benchmarks, including those for Word Sense Disambiguation (WSD), thanks to their ability in capturing context-sensitive semantic nuances. However, there is still little knowledge about their capabilities and po...

Cross-lingual embeddings represent the meaning of words from different languages in the same vector space. Recent work has shown that it is possible to construct such representations by aligning independently learned monolingual embedding spaces, and that accurate alignments can be obtained even without external bilingual data. In this paper we exp...

Cross-lingual embeddings represent the meaning of words from different languages in the same vector space. Recent work has shown that it is possible to construct such representations by aligning independently learned monolingual embedding spaces, and that accurate alignments can be obtained even without external bilingual data. In this paper we exp...

In this paper, we present WiC-TSV (\textit{Target Sense Verification for Words in Context}), a new multi-domain evaluation benchmark for Word Sense Disambiguation (WSD) and Entity Linking (EL). Our benchmark is different from conventional WSD and EL benchmarks for it being independent of a general sense inventory, making it highly flexible for the...

State-of-the-art methods for Word Sense Disambiguation (WSD) combine two different features: the power of pre-trained language models and a propagation method to extend the coverage of such models. This propagation is needed as current sense-annotated corpora lack coverage of many instances in the underlying sense inventory (usually WordNet). At th...