Web claims, seen as assertions shared on the web and eligible for fact-checking, are at the heart of online discourse. They have been studied extensively on a variety of downstream tasks such as fact-checking, claim retrieval, bias detection, argument mining, or viewpoint discovery. On the other hand, claims originating from scientific publications have also been the subject of several downstream NLP tasks. However, research carried out so far has yet to focus on scientific web claims, which are scientific claims made on the web (e.g., on social media and news articles). The process of detecting and fact-checking a claim from the web can be very different depending on whether the claim is scientific or not, thus making it crucial for the developed datasets, methods, and models to make a distinction between the two. With this work, we aim at understanding what makes this distinction necessary, by understanding the linguistic differences between scientific and non-scientific claims on the web, and the impact those differences have on existing downstream tasks. To do so, we manually annotate 1,524 web claims from established benchmarks for fact-checking-related tasks, and we run statistical tests to analyze and compare the linguistic features of each group. We find that scientific claims on the web use more analytical speech, but also use more sentiment-related speech, more expressions of physical motion, and have distinct parts of speech (PoS) and punctuation styles. We also conduct experiments showing that BERT-based language models perform worse on scientific web claims by up to 17 F1 points for several downstream tasks. To understand why, we develop a novel methodology to map predictive tokens of language models to explainable linguistic features and find that language models fail to detect a specific subset of predictive features of scientific web claims. We conclude by stating that language models aimed at studying scientific web claims ought to be trained on scientific web discourse, as opposed to being trained only on generic web discourse or only on scientific text from scientific publications.

Controversy refers to content attracting different point-of-views, as well as positive and negative feedback on a specific event, gathering users into different communities. Research on controversy led to two main categories of works: controversy detection/quantification and controversy explainability. When the former aims to quantify controversy on a topic, the latter aims to understand why a topic is controversial or not. This paper mainly contributes to the controversy explainability. We analyze topic discussions on Twitter from the community perspective to investigate the power of text in classifying tweets into the right community. We propose a SHAP-based pipeline to quantify impactful text features on predictions of three tweet classifiers. We also rely on the use of different text features namely BERT, TF − IDF, and LIWC. The results we obtain from both SHAP plots and statistical analysis show clearly significant impacts of some text features in classifying tweets.It also highlights the relevance of the study as well as the potential benefits of combining text and user interactions to quantify controversy.

Mental and physical health are strongly linked in a bidirectional relationship. Due to the stigma, ignorance, prejudice, fear, and many other reasons, there exists a large universal treatment gap for people with mental disorders. This could motivate those at-risk individuals to find their way into social networks, asking for information or emotional support. Language could provide a natural eyepiece for the study and detection of such at-risk individuals through their writings on social media platforms. In this paper, we consider the problem of detecting at-risk users with clear signs of depression, anorexia, self-harm, and suicidal thoughts. We introduce NCNL, a novel deep learning ensemble architecture that makes use of multiple noisy base learners in Negative Correlation Learning (NCL) configuration for text classification. NCNL is designed to be, backbone-independent, and we examine it with modern Transformer-based architectures. We evaluate our models on six different tasks for at-risk user detection and classification. Our models achieve significant improvements over existing state-of-the-art results reported for five out of the six tasks. Extensive experiments show how NCNL improves diversity over the classical conventional ensemble and the effect of using noisy base learners.

While neural networks (NN) have been successfully applied to many NLP tasks, the way they function is often difficult to interpret. In this article, we focus on binary text classification via NNs and propose a new tool, which includes a visualization of the decision boundary and the distances of data elements to this boundary. This tool increases the interpretability of NN. Our approach uses two innovative views: (1) an overview of the text representation space and (2) a local view allowing data exploration around the decision boundary for various localities of this representation space. These views are integrated into a visual platform, EBBE-Text, which also contains state-of-the-art visualizations of NN representation spaces and several kinds of information obtained from the classification process. The various views are linked through numerous interactive functionalities that enable easy exploration of texts and classification results via the various complementary views. A user study shows the effectiveness of the visual encoding and a case study illustrates the benefits of using our tool for the analysis of the classifications obtained with several recent NNs and two datasets.

Scientific topics, claims and resources are increasingly debated as part of online discourse, where prominent examples include discourse related to COVID-19 or climate change. This has led to both significant societal impact and increased interest in scientific online discourse from various disciplines. For instance, communication studies aim at a deeper understanding of biases, quality or spreading patterns of scientific information, whereas computational methods have been proposed to extract, classify or verify scientific claims using NLP and IR techniques. However, research across disciplines currently suffers from both a lack of robust definitions of the various forms of science-relatedness as well as appropriate ground truth data for distinguishing them. In this work, we contribute (a) an annotation framework and corresponding definitions for different forms of scientific relatedness of online discourse in tweets, (b) an expert-annotated dataset of 1261 tweets obtained through our labeling framework reaching an average Fleiss Kappa κ of 0.63, (c) a multi-label classifier trained on our data able to detect science- relatedness with 89% F1 and also able to detect distinct forms of scientific knowledge (claims, references). With this work, we aim to lay the foundation for developing and evaluating robust methods for analysing science as part of large-scale online discourse.

Data mining methods extract knowledge from huge amounts of data. Recently with the explosion of mobile technologies, a new type of data appeared. The resulting databases can be described as spatiotemporal data in which spatial information (e.g., the location of an event) and temporal information (e .g., the date of the event) are included. In this article, we focus on spatiotemporal patterns extraction from this kind of databases. These patterns can be considered as sequences representing changes of events localized in areas and its near surrounding over time. Two algorithms are proposed to tackle this problem: the first one uses \emph{a priori} strategy and the second one is based on pattern-growth approach. We have applied our generic method on two different real datasets related to: 1) pollution of rivers in France; and 2) monitoring of dengue epidemics in New Caledonia. Additionally, experiments on synthetic data have been conducted to measure the performance of the proposed algorithms.

BACKGROUND: The aim of this study was to develop an original method to extract sets of relevant molecular biomarkers (gene sequences) that can be used for class prediction and can be included as prognostic and predictive tools. MATERIALS AND METHODS: The method is based on sequential patterns used as features for class prediction. We applied it to classify breast cancer tumors according to their histological grade. RESULTS: We obtained very good recall and precision for grades 1 and 3 tumors, but, like other authors, our results were less satisfactory for grade 2 tumors. CONCLUSIONS: We demonstrated the interest of sequential patterns for class prediction of microarrays and we now have the material to use them for prognostic and predictive applications.

Today, many industrial companies must face challenges raised by maintenance. In particular, the anomaly detection problem is probably one of the most investigated. In this paper we address anomaly detection in new train data by comparing them to a source of normal train behavior knowledge, expressed as sequential patterns. To this end, fuzzy logic allows our approach to be both finer and easier to interpret for experts. In order to show the quality of our approach, experiments have been conducted on real and simulated anomalies.

Nowadays, the management of sequential patterns data becomes an increasing need in biological knowledge discovery processes. An important task in these processes is the restitution of the results obtained by using data mining methods. In a complex domain as biomedical, an efficient interpretation of the patterns without any assistance is difficult. One of the most common knowledge discovery process is clustering. But the application of clustering to gene sequential patterns is far from easy on biomedical data. In this paper, we introduce a new gene sequential patterns similarity function and summarization algorithm.

Dealing with digital data for mining novel knowledge is a non trivial task that has received much attention in the last years. However, it is still not easy to handle such data, especially when large volumes of values must be analyzed. In our work, we focus on biological data from DNA chips that biologists study in order to try and discover new gene correlations that could help understanding diseases like breast cancer. In this framework, we consider the values from the DNA microarrays, which convey the behavior of some genes, and we want to discover how these behaviors are correlated. This data are digital values that can be ordered and sorted. In previous work, sequential patterns like {(1 5)(2)} have been discovered, meaning that genes 1 and 5 have the same expression level followed by gene 2 that has a higher expression value. However, such data are very noisy and considering close values as ordered is often false. We thus consider here fuzzy rankings based on a fuzzy partition provided by the experts. Rules can then better characterize how genes are correlated.