Multilingual Detection Research Articles

Am I Hurt?: Evaluating Psychological Pain Detection in Hindi Text Using Transformer-based Models

The automated evaluation of pain is critical for developing effective pain management approaches that seek to alleviate pain while preserving patients’ functioning. Transformer-based models can aid in detecting pain from Hindi text data gathered from social media by leveraging their ability to capture complex language patterns and contextual information. By understanding the nuances and context of Hindi text, transformer models can effectively identify linguistic cues and sentiments and expressions associated with pain, enabling the detection and analysis of pain-related content present in social media posts. The purpose of this research is to analyze the feasibility of utilizing NLP techniques to automatically identify pain within Hindi textual data, providing a valuable tool for pain assessment in Hindi-speaking populations. The research showcases the HindiPainNet model, a deep neural network that employs the IndicBERT model, classifying the dataset into two class labels {pain, no_pain} for detecting pain in Hindi textual data. The model is trained and tested using a novel dataset, दर्द-ए-शायरी (pronounced as Dard-e-Shayari ), curated using posts from social media platforms. The results demonstrate the model's effectiveness, achieving an accuracy of 70.5%. This pioneer research highlights the potential of utilizing textual data from diverse sources to identify and understand pain experiences based on psychosocial factors. This research could pave the path for the development of automated pain assessment tools that help medical professionals comprehend and treat pain in Hindi-speaking populations. Additionally, it opens avenues to conduct further NLP-based multilingual pain detection research, addressing the needs of diverse language communities.

Navigating the Truth: Advanced Deep Learning Strategies Against the Spread of Digital Misinformation

This research presents a comprehensive secondary analysis on the utilization of deep learning technologies for the detection and mitigation of fake news. With the burgeoning challenge of digital misinformation compromising public discourse and trust, the study delves into an array of deep learning methodologies including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Transformer-based models like BERT and GPT for fake news detection. The analysis synthesizes existing methodologies, evaluates their effectiveness across various contexts, and critically assesses the ethical implications of automated detection, emphasizing the importance of privacy, bias, and censorship considerations. Key findings indicate that while deep learning models exhibit promise in identifying and mitigating fake news, continuous refinement and adaptation to evolving misinformation tactics are essential. The paper also underscores the role of secondary analysis in identifying research gaps, such as the need for multilingual detection capabilities and the integration of interdisciplinary approaches to enhance detection methodologies.

An Overview of the ADReSS-M Signal Processing Grand Challenge on Multilingual Alzheimer's Dementia Recognition Through Spontaneous Speech.

The ADReSS-M Signal Processing Grand Challenge was held at the 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023. The challenge targeted difficult automatic prediction problems of great societal and medical relevance, namely, the detection of Alzheimer's Dementia (AD) and the estimation of cognitive test scoress. Participants were invited to create models for the assessment of cognitive function based on spontaneous speech data. Most of these models employed signal processing and machine learning methods. The ADReSS-M challenge was designed to assess the extent to which predictive models built based on speech in one language generalise to another language. The language data compiled and made available for ADReSS-M comprised English, for model training, and Greek, for model testing and validation. To the best of our knowledge no previous shared research task investigated acoustic features of the speech signal or linguistic characteristics in the context of multilingual AD detection. This paper describes the context of the ADReSS-M challenge, its data sets, its predictive tasks, the evaluation methodology we employed, our baseline models and results, and the top five submissions. The paper concludes with a summary discussion of the ADReSS-M results, and our critical assessment of the future outlook in this field.

Multilingual Detection of Cyberbullying in Mixed Urdu, Roman Urdu, and English Social Media Conversations

Multilingual Detection of Cyberbullying in Mixed Urdu, Roman Urdu, and English Social Media Conversations

LLOD schema for Simplified Offensive Language Taxonomy in multilingual detection and applications

Abstract The goal of the paper is to present a Simplified Offensive Language (SOL) Taxonomy, its application and testing in the Second Annotation Campaign conducted between March-May 2023 on four languages: English, Czech, Lithuanian, and Polish to be verified and located in LLOD. Making reference to the previous Offensive Language taxonomic models proposed mostly by the same COST Action Nexus Linguarum WG 4.1.1 team, the number and variety of the categories underwent the definitional revision, and the present typology was tested in the annotation on the publicly available offensive language datasets of each of the four languages. The results of the annotation are presented and as they are contained within the accepted statistical values on the inter-annotator agreement in the SOL categories and their aspects, we propose this taxonomy as a core ontology which represents the encoding of the supported offensive languages and justify its use on new data in terms of a more universal Linguistic Linked Open Data (LLOD) schema.

Twitter Sentiment Analysis on Chatgpt Tweets

Abstract: This study investigates multilingual sentiment analysis within tweets on ChatGPT, an AI conversational model, employing Support Vector Machines (SVM) and BERT, an advanced language model. It aims to detect and classify emotions, including emoji identification, embedded within diverse messages across multiple languages on Twitter. By leveraging SVM's text classification and BERT's contextual understanding in various languages, the research delves into preprocessing techniques and feature engineering for sentiment analysis, encompassing multilingual and emoji detection. Furthermore, it explores the fusion of traditional SVM methods with BERT's state-of-the-art model for multilingual sentiment analysis, emphasizing emotion and emoji detection in AI-generated content on multilingual social media platforms like Twitter. This research yields insights into the successful detection of multilingual sentiment nuances and emotions, including emoji identification. It offers implications for advancing multilingual sentiment analysis in natural language processing across diverse linguistic contexts.

MMTD: A Multilingual and Multimodal Spam Detection Model Combining Text and Document Images

Spam detection has been a topic of extensive research; however, there has been limited focus on multimodal spam detection. In this study, we introduce a novel approach for multilingual multimodal spam detection, presenting the Multilingual and Multimodal Spam Detection Model combining Text and Document Images (MMTD). Unlike previous methods, our proposed model incorporates a document image encoder to extract image features from the entire email, providing a holistic understanding of both textual and visual content through a single image. Additionally, we employ a multilingual text encoder to extract textual features, enabling our model to process multilingual text content found in emails. To fuse the multimodal features, we employ a multimodal fusion module. Addressing the challenge of scarce large multilingual multimodal spam datasets, we introduce a new multilingual multimodal spam detection dataset comprising over 30,000 samples, which stands as the largest dataset of its kind to date. This dataset facilitates a rigorous evaluation of our proposed method. Extensive experiments were conducted on this dataset, and the performance of our model was validated using a five-fold cross-validation approach. The experimental results demonstrate the superiority of our approach, with our model achieving state-of-the-art performance, boasting an accuracy of 99.8% when compared to other advanced methods in the field.

Emotional Responses to Religious Conversion: Insights from Machine Learning

This study aims to understand the feelings of newly converted Muslims when they narrated their pre- and post-conversion using the Machine Learning model and qualitative approach. The data set analyzed in this paper comes from in-depth interviews with 12 mualaf/ newly converted Muslims from various backgrounds. All recorded interviews were transcribed and filtered to remove any unnecessary or misaligned data to ensure that the data was fully aligned with the interview questions. To analyze emotional changes, we utilize natural language processing (NLP) algorithms, which enable us to extract and interpret emotional content from textual data sources, such as personal narratives. The analysis was performed in Google Colab and utilizing XLM-EMO, a fine-tuned multilingual emotion detection model that detects joy, anger, fear, and sadness emotions from text. The model was chosen because it supports Bahasa, as our interview was conducted in Bahasa. Furthermore, the model also has the best accuracy amongst its competitors, namely LS-EMO and UJ-Combi. The model also has great performance, with the overall average Macro-F1s for XLM-RoBERTa-large, XLM-RoBERTa-base, and XLM-Twitter-base are .86, .81, and .84. Furthermore, two psychologists compared emotion detection results from the XLM-EMO model to the raw input data, and an inductive content analysis was performed. This approach allowed us to identify the reasoning behind the emotions deemed pertinent and intriguing for our investigation. This study showed that Sadness is the most dominant emotion, constituting 46.67% of the total emotions in the pre-conversion context. On the other hand, joy emerges as the most dominant, constituting a substantial proportion of 57.73% among the emotions analyzed from post-conversion emotions data. Understanding the positive impact of religious conversion on emotions may inform mental health interventions and incorporate spiritual or religious elements into therapeutic approaches for individuals struggling with emotional issues, guiding individuals undergoing religious conversion and emphasizing the potential emotional benefits.

XRBi-GAC: A hybrid deep learning framework for multilingual toxicity detection

Social media platforms allow people across the globe to share their thoughts and opinions and conveniently communicate with each other. Apart from various advantages of social media, it is also misused by a set of users for hate-mongering with toxic and offensive comments. The majority of the earlier proposed toxicity detection methods are primarily focused on the English language, but there is a lack of research on low-resource languages and multilingual text data. We propose an XRBi-GAC framework comprising XLM-RoBERTa, Bi-GRU with self-attention and capsule networks for multilingual toxic text detection. A loss function is also presented, which fuses the binary cross-entropy loss and focal loss to address the class imbalance problem. We evaluated the proposed framework on two datasets, namely, the Jigsaw Multilingual Toxic Comment dataset and HASOC 2019 dataset and achieved F1-score of 0.865 and 0.829, respectively. The results of the experiments show that the proposed framework has outperformed the state-of-the-art multilingual models XLM-RoBERTa and mBERT on both datasets, which shows the versatility and robustness of the proposed XRBi-GAC framework.

A modular approach for multilingual timex detection and normalization using deep learning and grammar-based methods

Detecting and normalizing temporal expressions is an essential step for many NLP tasks. While a variety of methods have been proposed for detection, best normalization approaches rely on hand-crafted rules. Furthermore, most of them have been designed only for English. In this paper we present a modular multilingual temporal processing system combining a fine-tuned Masked Language Model for detection, and a grammar-based normalizer. We experiment in Spanish and English and compare with HeidelTime, the state-of-the-art in multilingual temporal processing. We obtain best results in gold timex normalization, timex detection and type recognition, and competitive performance in the combined TempEval-3 relaxed value metric. A detailed error analysis shows that detecting only those timexes for which it is feasible to provide a normalization is highly beneficial in this last metric. This raises the question of which is the best strategy for timex processing, namely, leaving undetected those timexes for which is not easy to provide normalization rules or aiming for high coverage.

Automated Multilingual Detection of Pro-Kremlin Propaganda in Newspapers and Telegram Posts

The full-scale conflict between the Russian Federation and Ukraine generated an unprecedented amount of news articles and social media data reflecting opposing ideologies and narratives. These polarized campaigns have led to mutual accusations of misinformation and fake news, shaping an atmosphere of confusion and mistrust for readers worldwide. This study analyses how the media affected and mirrored public opinion during the first month of the war using news articles and Telegram news channels in Ukrainian, Russian, Romanian, French and English. We propose and compare two methods of multilingual automated pro-Kremlin propaganda identification, based on Transformers and linguistic features. We analyse the advantages and disadvantages of both methods, their adaptability to new genres and languages, and ethical considerations of their usage for content moderation. With this work, we aim to lay the foundation for further development of moderation tools tailored to the current conflict.

Zero-shot stance detection: Paradigms and challenges.

A major challenge in stance detection is the large (potentially infinite) and diverse set of stance topics. Collecting data for such a set is unrealistic due to both the expense of annotation and the continuous creation of new real-world topics (e.g., a new politician runs for office). Furthermore, stancetaking occurs in a wide range of languages and genres (e.g., Twitter, news articles). While zero-shot stance detection in English, where evaluation is on topics not seen during training, has received increasing attention, we argue that this attention should be expanded to multilingual and multi-genre settings. We discuss two paradigms for English zero-shot stance detection evaluation, as well as recent work in this area. We then discuss recent work on multilingual and multi-genre stance detection, which has focused primarily on non-zero-shot settings. We argue that this work should be expanded to multilingual and multi-genre zero-shot stance detection and propose best practices to systematize and stimulate future work in this direction. While domain adaptation techniques are well-suited for work in these settings, we argue that increased care should be taken to improve model explainability and to conduct robust evaluations, considering not only empirical generalization ability but also the understanding of complex language and inferences.

Multitask learning for multilingual intent detection and slot filling in dialogue systems

Dialogue systems are becoming an ubiquitous presence in our everyday lives having a huge impact on business and society. Spoken language understanding (SLU) is the critical component of every goal-oriented dialogue system or any conversational system. The understanding of the user utterance is crucial for assisting the user in achieving their desired objectives. Future-generation systems need to be able to handle the multilinguality issue. Hence, the development of conversational agents becomes challenging as it needs to understand the different languages along with the semantic meaning of the given utterance. In this work, we propose a multilingual multitask approach to fuse the two primary SLU tasks, namely, intent detection and slot filling for three different languages. While intent detection deals with identifying user’s goal or purpose, slot filling captures the appropriate user utterance information in the form of slots. As both of these tasks are highly correlated, we propose a multitask strategy to tackle these two tasks concurrently. We employ a transformer as a shared sentence encoder for the three languages, i.e., English, Hindi, and Bengali. Experimental results show that the proposed model achieves an improvement for all the languages for both the tasks of SLU. The multi-lingual multi-task (MLMT) framework shows an improvement of more than 2% in case of intent accuracy and 3% for slot F1 score in comparison to the single task models. Also, there is an increase of more than 1 point intent accuracy and 2 points slot F1 score in the MLMT model as opposed to the language specific frameworks.

Visual Object Detection with DETR to Support Video-Diagnosis Using Conference Tools

Real-time multilingual phrase detection from/during online video presentations—to support instant remote diagnostics—requires near real-time visual (textual) object detection and preprocessing for further analysis. Connecting remote specialists and sharing specific ideas is most effective using the native language. The main objective of this paper is to analyze and propose—through DEtection TRansformer (DETR) models, architectures, hyperparameters—recommendation, and specific procedures with simplified methods to achieve reasonable accuracy to support real-time textual object detection for further analysis. The development of real-time video conference translation based on artificial intelligence supported solutions has a relevant impact in the health sector, especially on clinical practice via better video consultation (VC) or remote diagnosis. The importance of this development was augmented by the COVID-19 pandemic. The challenge of this topic is connected to the variety of languages and dialects that the involved specialists speak and that usually needs human translator proxies which can be substituted by AI-enabled technological pipelines. The sensitivity of visual textual element localization is directly connected to complexity, quality, and the variety of collected training data sets. In this research, we investigated the DETR model with several variations. The research highlights the differences of the most prominent real-time object detectors: YOLO4, DETR, and Detectron2, and brings AI-based novelty to collaborative solutions combined with OCR. The performance of the procedures was evaluated through two research phases: a 248/512 (Phase1/Phase2) record train data set, with a 55/110 set of validated data instances for 7/10 application categories and 3/3 object categories, using the same object categories for annotation. The achieved score breaks the expected values in terms of visual text detection scope, giving high detection accuracy of textual data, the mean average precision ranging from 0.4 to 0.65.

Assessing the impact of OCR noise on multilingual event detection over digitised documents

Event detection is a crucial task for natural language processing and it involves the identification of instances of specified types of events in text and their classification into event types. The detection of events from digitised documents could enable historians to gather and combine a large amount of information into an integrated whole, a panoramic interpretation of the past. However, the level of degradation of digitised documents and the quality of the optical character recognition (OCR) tools might hinder the performance of an event detection system. While several studies have been performed in detecting events from historical documents, the transcribed documents needed to be hand-validated which implied a great effort of human expertise and manual labour-intensive work. Thus, in this study, we explore the robustness of two different event detection language-independent models to OCR noise, over two datasets that cover different event types and multiple languages. We aim at analysing their ability to mitigate problems caused by the low quality of the digitised documents and we simulate the existence of transcribed data, synthesised from clean annotated text, by injecting synthetic noise. For creating the noisy synthetic data, we chose to utilise four main types of noise that commonly occur after the digitisation process: Character Degradation, Bleed Through, Blur, and Phantom Character. Finally, we conclude that the imbalance of the datasets, the richness of the different annotation styles, and the language characteristics are the most important factors that can influence event detection in digitised documents.

Development of HAMOD: a High Agreement Multi-lingual Outlier Detection dataset

Development of HAMOD: a High Agreement Multi-lingual Outlier Detection dataset

Cross corpus multi-lingual speech emotion recognition using ensemble learning

Receiving an accurate emotional response from robots has been a challenging task for researchers for the past few years. With the advancements in technology, robots like service robots interact with users of different cultural and lingual backgrounds. The traditional approach towards speech emotion recognition cannot be utilized to enable the robot and give an efficient and emotional response. The conventional approach towards speech emotion recognition uses the same corpus for both training and testing of classifiers to detect accurate emotions, but this approach cannot be generalized for multi-lingual environments, which is a requirement for robots used by people all across the globe. In this paper, a series of experiments are conducted to highlight an ensemble learning effect using a majority voting technique for cross-corpus, multi-lingual speech emotion recognition system. A comparison of the performance of an ensemble learning approach against traditional machine learning algorithms is performed. This study tests a classifier’s performance trained on one corpus with data from another corpus to evaluate its efficiency for multi-lingual emotion detection. According to experimental analysis, different classifiers give the highest accuracy for different corpora. Using an ensemble learning approach gives the benefit of combining all classifiers’ effect instead of choosing one classifier and compromising certain language corpus’s accuracy. Experiments show an increased accuracy of 13% for Urdu corpus, 8% for German corpus, 11% for Italian corpus, and 5% for English corpus from with-in corpus testing. For cross-corpus experiments, an improvement of 2% when training on Urdu data and testing on German data and 15% when training on Urdu data and testing on Italian data is achieved. An increase of 7% in accuracy is obtained when testing on Urdu data and training on German data, 3% when testing on Urdu data and training on Italian data, and 5% when testing on Urdu data and training on English data. Experiments prove that the ensemble learning approach gives promising results against other state-of-the-art techniques.

Multilingual stance detection in social media political debates

Stance Detection is the task of automatically determining whether the author of a text is in favor, against, or neutral towards a given target. In this paper we investigate the portability of tools performing this task across different languages, by analyzing the results achieved by a Stance Detection system (i.e. MultiTACOS) trained and tested in a multilingual setting.First of all, a set of resources on topics related to politics for English, French, Italian, Spanish and Catalan is provided which includes: novel corpora collected for the purpose of this study, and benchmark corpora exploited in Stance Detection tasks and evaluation exercises known in literature. We focus in particular on the novel corpora by describing their development and by comparing them with the benchmarks. Second, MultiTACOS is applied with different sets of features especially designed for Stance Detection, with a specific focus to exploring and combining both features based on the textual content of the tweet (e.g., style and affective load) and features based on contextual information that do not emerge directly from the text. Finally, for better highlighting the contribution of the features that most positively affect system performance in the multilingual setting, a features analysis is provided, together with a qualitative analysis of the misclassified tweets for each of the observed languages, devoted to reflect on the open challenges.

#europehappinessmap: A Framework for Multi-Lingual Sentiment Analysis via Social Media Big Data (A Twitter Case Study)

The growth and popularity of social media platforms have generated a new social interaction environment thus a new collaboration and communication network among individuals. These platforms own tremendous amount of data about users’ behaviors and sentiments since people create, share or exchange their information, ideas, pictures or video using them. One of these popular platforms is Twitter, which via its voluntary information sharing structure, provides researchers data potential of benefit for their studies. Based on Twitter data, in this study a multilingual sentiment detection framework is proposed to compute European Gross National Happiness (GNH). This framework consists of a novel data collection, filtering and sampling method, and a newly constructed multilingual sentiment detection algorithm for social media big data, and tested with nine European countries (United Kingdom, Germany, Sweden, Turkey, Portugal, The Netherlands, Italy, France and Spain) and their national languages over a six year period. The reliability of the data is checked with peak/troughs comparison for special days from Wikipedia news lists. The validity is checked with a group of correlation analyses with OECD Life Satisfaction survey reports’, Euro-Dollar and other currency exchanges, and national stock market time series data. After validity and reliability confirmations, the European GNH map is drawn for six years. The main problem addressed is to propose a novel multilingual social media sentiment analysis framework for calculating GNH for countries and change the way of OECD type organizations’ survey and interview methodology. Also, it is believed that this framework can serve more detailed results (e.g., daily or hourly sentiments of society in different languages).

English-Persian Plagiarism Detection based on a Semantic Approach

Plagiarism which is defined as “the wrongful appropriation of other writers’ or authors’ works and ideas without citing or informing them” poses a major challenge to knowledge spread publication. Plagiarism has been placed in four categories of direct, paraphrasing (rewriting), translation, and combinatory. This paper addresses translational plagiarism which is sometimes referred to as cross-lingual plagiarism. In cross-lingual translation, writers meld a translation with their own words and ideas. Based on monolingual plagiarism detection methods, this paper ultimately intends to find a way to detect cross-lingual plagiarism. A framework called Multi-Lingual Plagiarism Detection (MLPD) has been presented for cross-lingual plagiarism analysis with ultimate objective of detection of plagiarism cases. English is the reference language and Persian materials are back translated using translation tools. The data for assessment of MLPD were obtained from English-Persian Mizan parallel corpus. Apache’s Solr was also applied to record the creep of the documents and their indexation. The accuracy mean of the proposed method revealed to be 98.82% when employing highly accurate translation tools which indicate the high accuracy of the proposed method. Also, Google translation service showed the accuracy mean to be 56.9%. These tests demonstrate that improved translation tools enhance the accuracy of the proposed method.