Language Technology Research Articles

Ethical considerations related to personal data collection and reuse: trust and transparency in language and speech technologies

Ethical considerations related to personal data collection and reuse: trust and transparency in language and speech technologies

Recoding of a society

The undercarriage of Afrikaner nationalist mobilisation was assembled by a small group of language engineers during the opening decades of the twentieth century. Particularly influential within this broad literary project were the Afrikaans clerics who clustered around the Zuid-Afrikaanse Bijbel Vereniging (ZABV, later N.G. Kerk Uitgewers van Suid-Afrika), the official publishing wing of the Dutch Reformed Church (DRC). Propped up by this platform, DRC writers contributed to the linguistic shape of Afrikaans and crucially steered the affixed matter of nation building along a defined religious route in the image of a Christian-National Afrikanerdom. By the start of 1940 a sociological function within the operation of the ZABV lent itself to the nationalist agenda of capturing the consciousness of Afrikaner society. This notably contributed to the creation of an indoctrinated and self-concealed Afrikaner bloc based on the perpetuated myth of an unquestioned and divinely affirmed national uniqueness.

New Media Advertising Information Search Method Based on XML Technology

The sheer volume and complexity of new media advertising information pose significant challenges for users in searching and filtering advertisements. To enhance the match between search content and keywords, this paper proposes a multifunctional search method for new media advertising information based on eXtensible Markup Language (XML) technology. The collected user attribute data are organized into a set, and a membership function for user attributes is established. An information certainty index and a query certainty index are constructed to obtain the integrated attributes of new media advertisements and classify their attributes. The average value of the constant threshold element is calculated, and an attraction model of new media advertisements for random Internet users is constructed. The similarity between text information and query methods is determined, along with the calculation of browsing duration and frequency in any given area. According to the search model for new media advertising information, the sorted new media advertising information documents are merged to obtain search results. Experimental results demonstrate that the introduction of interactive elements enhances attraction, and dynamic new media advertisements exhibit stronger attraction compared to static advertisements. The proposed method shows significant advantages in search efficiency, achieving over 90% search efficiency, with the match between search content and keywords consistently remaining above 98%, the user satisfaction is higher, the system response time is shorter, the results can be obtained faster, and the search efficiency is improved.

A knowledge-Aware NLP-Driven conversational model to detect deceptive contents on social media posts

The widespread dissemination of deceptive content on social media presents a substantial challenge to preserving authenticity and trust. The epidemic growth of false news is due to the greater use of social media to transmit news, rather than conventional mass media such as newspapers, magazines, radio, and television. Humans' incapacity to differentiate among true and false facts exposes fake news as a threat to logical truth, democracy, journalism, and government credibility. Using combination of advanced methodologies, Deep learning (DL) methods, and Natural Language Processing (NLP) approaches, researchers and technology developers attempt to make robust systems proficient in discerning the subtle nuances that betray deceptive intent. Analysing conversational linguistic patterns of misleading data, these techniques’ purpose to progress the resilience of social platforms against the spread of deceptive content, eventually contributing to an additional informed and trustworthy online platform. This paper proposed a Knowledge-Aware NLP-Driven AlBiruni Earth Radius Optimization Algorithm with Deep Learning Tool for Enhanced Deceptive Content Detection (BER-DLEDCD) algorithm on Social Media. The purpose of the BER-DLEDCD system is to identify and classify the existence of deceptive content utilizing NLP with optimal DL model. In the BER-DLEDCD technique, data pre-processing takes place to change the input data into compatible format. Furthermore, the BER-DLEDCD approach applies hybrid DL technique encompassing Convolutional Neural Network with Long Short-Term Memory (CNN-LSTM) methodology for deceptive content detection. Moreover, the BER approach has been deployed to boost hyperparameter choice of the CNN-LSTM technique which leads to enhanced detection performance. The simulation outcome of the BER-DLEDCD system has been examined employing benchmark database. The extensive outcomes stated the BER-DLEDCD system achieved excellent performance with the accuracy of 94 %, 94.83 % precision, 94.30 % F-score with other recent approaches.

Designing Semantic Tutoring Maps for Enhancing Peer Learning through Lesson Preparation

Advancements in Natural Language Processing technology present significant opportunities to enhance student learning through peer tutoring. While there is substantial research on the effectiveness of peer tutoring in learning science, few studies focus on the lesson preparation stage of peer tutoring. This study utilizes Semantic Web technologies to organize tutoring maps, aiming to develop an innovative learning artifact to address this research gap. Adopting a design-based research methodology, this study first provides a literature review regarding the application of semantic technology in education and peer tutoring preparation. Subsequently, a tutoring map application based on a semantic database is developed, encompassing the organization and storage of tutoring lesson plans using Resource Description Framework (RDF) technology, and the querying and feedback on lesson plan quality based on SPARQL Protocol and RDF Query Language (SPARQL) technology. This paper describes the theoretical basis and design rationale by which student tutors’ learning is improved through structured tutoring preparation. This includes enhancing cognitive and meta-cognitive abilities in designing learning activities, raising questions, anticipating answers, and providing guidance responses to incorrect answers. The paper concludes with recommendations for educational researchers and practitioners. Future research will focus on empirical studies of peer tutoring in scientific fields to verify the validity and adaptability of the design, further facilitating peer learning by tutoring and collective knowledge building.

From greatest simplicity to full power

AbstractWhile language processing services are key assets for the science and technology of language, the possible ways under which they may be made available to the widest range of their end users are critical to support an Open Science policy for this scientific domain. Although providing such processing services under some web-based interface, at large, offers itself as an immediate and cogent response to that challenge, turning this view into an effective access to language processing services is an undertaking deserving a clear conceptual direction and a corresponding robust empirical validation. Based on an extensive overview of major undertakings towards making language processing tools available and on the design principles worked out and implemented in the PORTULAN CLARIN infrastructure, in this paper we advocate for a Research-Infrastructure-as-a-Service (RIaaS) model. This model unleashes accessibility to language processing services in as many web-based interface modalities as the current stage of technological development permits to support, in order to serve as many types of end users as possible, from IT developers to Digital Humanities researchers, and including citizen scientists, teachers, students and digital artists among many others.

The application of GIS technology in building a multivariate taphonomic profile for improving PMI estimations in Greece.

Environmental conditions highly affect decomposition rates and therefore a forensic practitioner should consider context-specific information when estimating the post mortem interval (PMI). Traditional methods of collecting environmental data, however, are time-consuming and often impractical for large-scale studies or routine forensic investigations. This study developed an automated computer method by employing the technology of geographic information systems (GIS) and Python programming language to provide contextual information for bodies found outdoors in Greece. The generated coding script underwent testing on 95 bodies in various stages of decomposition, which were examined between the years 1999 and 2022 at the National and Kapodistrian University of Athens and the Forensic Medical Service of Thessaloniki. Using ArcGIS Pro software and publicly available online data, a multilayer map was developed. Individual layers included high-resolution aerial images and data on the European Nature Information System ecosystem type, the Köppen-Geiger climatic type, the population density, the elevation, and the slope. Additionally, 99 national weather stations and their corresponding meteorological data were integrated. By leveraging the geographical coordinates of the recovery site of each case and information about the decedent's disappearance and recovery dates, this script automatically generates details from each of the above layers. Additionally, it calculates the accumulated degree days (ADD) and accumulated humidity days (AHD) values by extracting data from the nearest weather station. The GIS-based approach enables rapid, objective, and reproducible taphonomic profile construction, which can greatly improve the reliability of PMI estimations. By utilizing this method, forensic practitioners can accurately evaluate environmental effects on decomposition, thus standardizing taphonomic profiling globally.

Multilinguality and LLOD: A survey across linguistic description levels

Limited accessibility to language resources and technologies represents a challenge for the analysis, preservation, and documentation of natural languages other than English. Linguistic Linked (Open) Data (LLOD) holds the promise to ease the creation, linking, and reuse of multilingual linguistic data across distributed and heterogeneous resources. However, individual language resources and technologies accommodate or target different linguistic description levels, e.g., morphology, syntax, phonology, and pragmatics. In this comprehensive survey, the state-of-the-art of multilinguality and LLOD is being represented with a particular focus on linguistic description levels, identifying open challenges and gaps as well as proposing an ideal ecosystem for multilingual LLOD across description levels. This survey seeks to contribute an introductory text for newcomers to the field of multilingual LLOD, uncover gaps and challenges to be tackled by the LLOD community in reference to linguistic description levels, and present a solid basis for a future best practice of multilingual LLOD across description levels.

Machine Translation and NLP Tools: Developing and Refining Language Technologies for African Languages

Machine Translation (MT) and Natural Language Processing (NLP) tools have significantly evolved over the past few decades, leading to notable advancements in the field of language technology. However, most of these developments have been concentrated on languages with abundant data, such as English, Chinese, and European languages. African languages, with their unique linguistic structures and vocabularies, remain underrepresented in the MT and NLP landscape (Joshi et al., 2020). This article explores the challenges and opportunities in creating and refining machine translation models and NLP tools tailored to the unique structures of African languages. It discusses the specific linguistic features that set African languages apart, reviews the current state of MT and NLP tools for these languages, and outlines strategies for developing more effective models that cater to the diverse linguistic landscape of Africa.

Transnational Research Infrastructure

This paper discusses best practices in the development and operation of the network of Knowledge centres within the framework of the transnational and cross-disciplinary CLARIN Knowledge Infrastructure. A Knowledge centre is an institution or a collective of institutions which “have agreed to share their knowledge and expertise with others” (Branco et al. 2023), which has been certified by Common Language Resources and Technology Infrastructure CLARIN, and which has an established website providing the users with information on the offered services and contact forms. Currently, there are 28 Knowledge centres from different countries covering a wide range of topics, offering various research and training related services and organised as either a virtual centre of one or several institutional partners in one country or distributed across multiple countries. This paper aims to present this rich diversity of the research infrastructure and its state-of-the-art by reviewing existing Knowledge centres and their areas of expertise, describing the activities initiated by CLARIN to strengthen K-centre networking, showcasing different types of K-centres, and presenting the operation and collaboration practices shared by K-centres in their annual reports.

SWENER-1800

Named entity recognition (NER) is the process of automatically identifying persons, places, organisations and other name-like entities in text, in order to perform natural language processing tasks such as automatic extraction of metadata from text, anonymisation/pseudonymisation of sensitive personal data, or as a preprocessing step for linking different terms describing the same entity to a single reference. While NER is a mature language technology, it is generally lacking for historical language varieties. We describe our work on compiling SWENER-1800, a large (half a million words) reference corpus of historical Swedish texts, covering the time period from the first half of the 18th century until about 1900, and manually annotating it with named entity types identified as significant for this time period, as well as with sentence boundaries, notoriously difficult to recognise automatically in historical text. This corpus can then be used to train and evaluate NER systems and sentence segmenters for historical Swedish text. An additional concrete contribution from this work is a manual for annotation of named entities in historical Swedish.

Language Barriers, Technology Adoption and Productivity: Evidence from Agriculture in India

Abstract We study the effect of language barriers on the ability of farmers to access information about agricultural technologies in rural areas of India. We use the introduction of government-sponsored call centers (Kisan Call Centers) which offer agricultural advice in the official language of each Indian state. For identification, we compare geographically contiguous areas that sit across state borders, and exploit differences in the language spoken by farmers and call center advisors. We document that language barriers limit the adoption of modern agricultural technologies – such as high-yielding variety seeds – and negatively affect crop yields.

Interactive probes: Towards action-level evaluation for dialogue systems

Measures of ‘humanness’, ‘coherence’ or ‘fluency’ are the mainstay of dialogue system evaluation, but they don’t target system capabilities and rarely offer actionable feedback. Reviewing recent work in this domain, we identify an opportunity for evaluation at the level of action sequences, rather than the more commonly targeted levels of whole conversations or single responses. We introduce interactive probes, an evaluation framework inspired by empirical work on social interaction that can help to systematically probe the capabilities of dialogue systems. We sketch some first probes in the domains of tellings and repair, two sequence types ubiquitous in human interaction and challenging for dialogue systems. We argue interactive probing can offer the requisite flexibility to keep up with developments in interactive language technologies and do justice to the open-endedness of action formation and ascription in interaction.

IAT/ML: a metamodel and modelling approach for discourse analysis

Language technologies are gaining momentum as textual information saturates social networks and media outlets, compounded by the growing role of fake news and disinformation. In this context, approaches to represent and analyse public speeches, news releases, social media posts and other types of discourses are becoming crucial. Although there is a large body of literature on text-based machine learning, it tends to focus on lexical and syntactical issues rather than semantic or pragmatic. Being useful, these advances cannot tackle the nuanced and highly context-dependent problems of discourse evaluation that society demands. In this paper, we present IAT/ML, a metamodel and modelling approach to represent and analyse discourses. IAT/ML focuses on semantic and pragmatic issues, thus tackling a little researched area in language technologies. It does so by combining three different modelling approaches: ontological, which focuses on what the discourse is about; argumentation, which deals with how the text justifies what it says; and agency, which provides insights into the speakers’ beliefs, desires and intentions. Together, these three modelling approaches make IAT/ML a comprehensive solution to represent and analyse complex discourses towards their understanding, evaluation and fact checking.

Advancing Sustainable Cyber-Physical System Development with a Digital Twins and Language Engineering Approach: Smart Greenhouse Applications

In recent years, the integration of Internet of Things technologies in smart agriculture has become critical for sustainability and efficiency, to the extent that recent improvements have transformed greenhouse farming. This study investigated the complexity of IoT architecture in smart greenhouses by introducing a greenhouse language family (GreenH) that comprises three domain-specific languages designed to address various tasks in this domain. The purpose of this research was to streamline the creation, simulation, and monitoring of digital twins, an essential tool for optimizing greenhouse operations. A three-stage methodology was employed to develop the GreenH DSLs, a detailed metamodel for enhanced smart monitoring systems. Our approach used high-level metamodels and extended Backus–Naur form notation to define the DSL syntax and semantics. Through a comprehensive evaluation strategy and a selected language usability metrics, the expressiveness, consistency, readability, correctness, and scalability of the DSL were affirmed, and areas for usability improvement were highlighted. The findings suggest that GreenH languages hold significant potential for advancing digital twin modeling in smart agriculture. Future work should be aimed at refining usability and extending its application range. The anticipated integration with additional model-drive engineering and code generation tools will improve interoperability and contribute to digital transformation in the smart greenhouse domain and promote more sustainable food production systems.

Impacts of LMS Socio-Linguistic and Psychometric Factors on Students' English and Translation Proficiency and Communicative Competence: A Paradigm Shift During COVID-19 Pandemic

This quantitative study investigated the impacts of Learning Management Systems’ (LMS) socio-linguistic and psychometric factors on students' English and translation proficiency and communicative competence. The participants were enrolled at the College of Languages and Translation, Imam Mohammed Ibn Saud Islamic University (IMSIU), Riyadh, Saudi Arabia. The results highlighted the positive experiences students had with the LMS platforms, with a notable frequency of (79.6%). Additionally, a significant percentage (86.2%) affirmed the effectiveness of these platforms in aiding them in achieving academic objectives. However, (52%) indicated that students consistently or often faced challenges when using technology. The findings highlighted improvements in various areas, including linguistic abilities (71.7%), translation skills (73.7%), social skills (62.5%), and psychometric abilities (70.3%), emphasizing the role of online environments in fostering effective communicative skills. The study also revealed the benefits of language and translation technology for students, with (84.1%) using computer-assisted translation, (82.9%) engaging in dubbing and voice translation, and (75.6%) utilizing language audio/video resources. Positive impacts on proficiency and communicative competence were evident, with frequencies of (74.4%) and (69.1%), respectively. These findings underscore the positive influence of technology on language and translation learning, emphasizing communicative approaches that enable students to develop competencies based on language socialization and real-time translation practices. The study recommends ongoing technology training for students and advocates for the use of psychometric tools embedded in LMS to encourage self-assessment and independent learning. The study's significance lies in understanding the intricate relationship between cognition, educational technology, and language and translation learning, particularly in the context of higher education.

Can AI Make a Case? AI Vs. Lawyer in The Dutch Legal Context

The integration of AI, specifically GPT-4, into the legal field is a subject of both potential promise and intricate challenges. This thesis delves into the transformational possibilities of AI within the Dutch legal context, examining not only the quality and persuasiveness of AI-generated legal argumentation but also its competence in information retrieval, as measured by models’ ability to spot relevant legal issues. An experiment was conducted with 25 legal professionals, using a real-world Dutch case with the purpose to assess GPT-4’s capabilities against that of a human lawyer. To enable GPT-4 to handle case documents, the author first performed so-called co-reference resolution to remove ambiguities. Given the token limitations of GPT-4, a so-called prompt reducer technique was used to compress the text, retaining essential information. The above methods produced a coherent and full case summary within GPT-4’s token constraints. This case summary, together with original lawyer’s letter (denoted as Text A) was fed to ChatGPT-4 to obtain its AI-written alternative (Text B). The study subjects were presented with a case summary as well as both texts and asked for their preferences. The outcome of the experiment is as follow: 80% of participants chose the AI’s composed legal document, demonstrating a strong preference for both its linguistic competencies as well its ability to spot relevant legal issues. This preference for GPT-4 writing is very consistent among genders, age groups and professions surveyed. Contextualising these findings within the broader implications for legal practice, the thesis explores potential benefits including increased access to justice and transformation of certain legal procedures. Insights and recommendations are offered for legal professionals, considering the technological evolution and ethical considerations inherent in AI integration. Acknowledging the need for further exploration, the study recognises its own limitations and encourages replication to solidify the understanding of AI’s transformative role in the legal realm. GPT-4, AI, AI vs Lawyer, Legal Writing, Language Technology, LLM, Dutch Legal Context

Dreaming Fish, Becoming Fish: Animating Modern Mythology and Zhuangzian Philosophy in the Fantasy Film Big Fish & Begonia

Abstract This paper analyzes the 2016 Chinese animation film Big Fish & Begonia and its precursor, a 2004 Flash animated short video, in conjunction with Zhuangzi’s philosophy of becoming and transformation. It traces the origin of the film in the auteur’s dream experience and the development of cinematic language and animation technology from the initial short video to the full-feature film. Adopting a form closely associated with the fantastical and speculative traditions of Chinese mythology, Big Fish & Begonia represents animal becomings through the apparatus of the dream, as both crucial content and form, and the formal characteristics of animation. Both the dream and the animation forms allow the expression of desires to break through the ontological and epistemological constraints of the human subject and to render the physical form unstable. By engaging with key passages from Zhuangzi that discuss dreams, knowledge, “free and easy wandering,” and “becoming/transformation,” the paper asserts Zhuangzi’s continued influence in contemporary Chinese animation. The film offers new ways of conceptualizing and addressing modern environmental crises while aligning itself with the cultural–political agenda to create “authentic” Chinese stories.

AI generates covertly racist decisions about people based on their dialect

Hundreds of millions of people now interact with language models, with uses ranging from help with writing1,2 to informing hiring decisions3. However, these language models are known to perpetuate systematic racial prejudices, making their judgements biased in problematic ways about groups such as African Americans4–7. Although previous research has focused on overt racism in language models, social scientists have argued that racism with a more subtle character has developed over time, particularly in the United States after the civil rights movement8,9. It is unknown whether this covert racism manifests in language models. Here, we demonstrate that language models embody covert racism in the form of dialect prejudice, exhibiting raciolinguistic stereotypes about speakers of African American English (AAE) that are more negative than any human stereotypes about African Americans ever experimentally recorded. By contrast, the language models’ overt stereotypes about African Americans are more positive. Dialect prejudice has the potential for harmful consequences: language models are more likely to suggest that speakers of AAE be assigned less-prestigious jobs, be convicted of crimes and be sentenced to death. Finally, we show that current practices of alleviating racial bias in language models, such as human preference alignment, exacerbate the discrepancy between covert and overt stereotypes, by superficially obscuring the racism that language models maintain on a deeper level. Our findings have far-reaching implications for the fair and safe use of language technology.

A deaf-centred art-science approach to community engagement with sign language technologies

Sign Language Machine Translation (SLMT) is an emerging technology that will primarily impact deaf and hard of hearing communities. Historically, these technologies have been developed by hearing scientists, without input from deaf researchers or community representatives. SignON was a three-year project exploring the application of SLMT, and using a co-creation approach which was led by the European Union of the Deaf. Here we describe an art-science method to engage deaf, hard of hearing, and hearing audiences with SLMT through theatre and performance. We also reflect on the insights, perspectives, and feedback shared by audience members.