Search Results (15)

Background: There is an ongoing project to create an international Thyroid Imaging Reporting And Data System (I-TIRADS) to harmonize the terminology of guidelines for reporting thyroid ultrasonography. As artificial intelligence (AI) has been gaining increasing attention also in the thyroid field, achieving solid information about the consistency of AI in interpreting the TIRADS terminology is relevant before the I-TIRADS is published. The present study aimed to examine the issue of AI when interpreting the TIRADS terminology to describe thyroid nodules (TNs). Methods: Three TIRADSs from the USA (ACR-TIRADS), Europe (EU-TIRADS), and Asia (K-TIRADS) were considered. The most popular AI, such as ChatGPT, was tested. All possible combinations of terms of the three TIRADSs were performed. Results: 2592 cases were included. With the ACR-TIRADS lexicon, there was a slightly significant difference between systems (p = 0.0494) which was attributed to variations between ACR- and EU-TIRADS (p = 0.0099). With the EU-TIRADS lexicon, there was a significant difference between systems (p < 0.0001) with a significant result between EU- and ACR-TIRADS (p = 0.0003). Using the K-TIRADS terminology, no significant difference was observed (p = 0.7954). The intraobserver agreement of ChatGPT was moderate with the best values (from 0.55 to 0.60) with the K-TIRADS lexicon. Conclusions: ChatGPT interprets the TIRADS lexicon but with variations when it is asked to assess TNs according to one TIRADS using the terminology of another TIRADS. Clinical operators as well as patients should also be aware of these novel data. Full article

Artificial intelligence (AI) is a cutting-edge and revolutionary technology in computer science that has the potential to completely transform a wide range of disciplines, including the social sciences, the arts, and the humanities. Therefore, since its significance has been recognized in engineering and medicine, history, literature, paleography, and archaeology have recently embraced AI as new opportunities have arisen for preserving ancient manuscripts. Acknowledging the importance of digitizing archival documents, this paper explores the use of advanced technologies during this process, showing how these are employed at each stage and how the unique challenges inherent in past scripts are addressed. Our study is based on Cyril’s Lexicon, a Byzantine-era dictionary of great historical and linguistic significance in Greek territory. Full article

The rapid expansion of artificial intelligence (AI) in the workplace has reshaped job roles, yet its effect on employee satisfaction in AI-specific positions remains underexplored. We curated n = 1500 Glassdoor reviews of AI professionals (70% dated 2018–2025). We applied lexicon-based sentiment analysis (TextBlob) alongside R-driven statistical modelling to (1) quantify star ratings, (2) compare narrative sentiment with numerical scores, and (3) track annual sentiment trends from 2018 to 2025. AI specialists report high overall satisfaction (mean = 4.24) and predominantly positive sentiment (80.7%), although perceptions of leadership quality are consistently lower (mean = 3.94). Our novel AI-focused dataset and dual-analysis pipeline offer a scalable foundation for real-time HR dashboards. These tools can help organisations anticipate workforce needs, target leadership development, and implement ethical, data-driven AI practices to sustain employee well-being. Full article

Limited research has systematically compared the detailed sensory profiles of commercially available gluten-containing (C) and gluten-free (F) cookies using trained panelists. This study aimed to develop a comprehensive sensory lexicon for C and F chocolate chip cookies and identify key sensory attributes that differentiate them. Seven professionally trained panelists created a lexicon of 33 attributes spanning aroma, flavor, basic taste, texture, and residual property. Using this lexicon, a descriptive analysis was conducted on 12 C and 12 F cookie samples. Multivariate analysis of variance revealed significant differences between the two groups across the 33 sensory attributes (p < 0.05). A mixed model analysis showed that C cookies had higher intensities of chocolate-related and sweet aroma complex notes, while F cookies exhibited stronger nutty, artificial, and off-note flavors. In terms of texture, F cookies were higher in toothpack and powdery mouthcoat, while C cookies displayed more melt-in-mouth characteristics. Principal component analysis and agglomerative hierarchical clustering revealed three distinct clusters of test samples within both crispy and chewy cookie types, with some F cookies closely aligning with C profiles. These findings, along with the developed lexicon, provide a valuable foundation for enhancing the sensory appeal and quality of gluten-free chocolate chip cookies. Full article

Previous studies have shown that language learners’ auditory word recognition behavior provides evidence for independent contributions of perceptual and phonolexical representations, and learners’ patterns of auditory word recognition have been characterized as resulting from “fuzziness” or “imprecision” associated with these representations. More recently, it has been argued that representational “fuzziness” may in fact take various forms (e.g., neutralized, precise, ambiguous). The purpose of the present study is to further build on this line of work by elaborating additional logically possible scenarios by crossing larger sets of logically possible types of perceptual and phonolexical representational precision/imprecision, as an exercise in exploring the empirical and theoretical implications of our characterizations of representational fuzziness in language learners. We collect new empirical data for the purpose of demonstrating how we might evaluate auditory word recognition performance relative to this fuller set of predicted scenarios. We computed the set of hypothesized scenarios by crossing possible perceptual and lexical representations. We crossed four possible perceptual representations (NeutralizedC + NeutralizedV, NeutralizedC + PreciseV, PreciseC + NeutralizedV, or PreciseC + PreciseV) and six possible phonolexical representations (Neutralized, Ambiguous, Not X, Precise, Fuzzy Word, or Word Length), for a total of 24 scenarios, each accompanied by a set of predictions with respect to accuracy on an auditory word–picture matching test. We interpret the group and individual performance relative to these scenarios with the ultimate aim of better understanding the implications of our assumptions about the nature of perceptual and phonolexical representations relative to observed patterns of learner behavior. Our hope is that in computing this factorial typology of logically possible scenarios and demonstrating a starting point for how we might empirically evaluate its predictions, we set the stage for future research to refine the hypothesis space through empirical studies of auditory word processing in language learners. Full article

The adoption of emerging technologies in healthcare systems plays a crucial part in anti-obesity initiatives. COVID-19 has intensified the Body Mass Index (BMI) discourses in AI (Artificial Intelligence)-powered social media. However, few studies have reported on the influence of digital content on obesity prevention policies. Understanding the nature and forums of obese metaphors in social media is the first step in policy intervention. The purpose of this paper is to understand the mutual influence between obesity and COVID-19 and determine its policy implications. This paper analyzes the public responses to obesity using Twitter data collected during the COVID-19 pandemic. The emotional nature of tweets is analyzed using the NRC lexicon. The results show that COVID-19 significantly influences perceptions of obesity; this indicates that existing public health policies must be revisited. The study findings delineate prerequisites for obese disease control programs. This paper provides policy recommendations for improving social media interventions in health service delivery in order to prevent obesity. Full article

Nowadays, the social dimension of product sustainability is increasingly in demand, however, industrial designers struggle to pursue it much more than the environmental or economic one due to their unfamiliarity in correlating design choices with social impacts. In addition, this gap is not filled even by the supporting methods that have been conceived to only support specific areas of application. To fill this gap, this study proposed a method to support social failure mode and effect analysis (SFMEA), though the automatic failure determination, based on the use of a chatbot (i.e., an artificial intelligence (AI)-based chat). The method consists of 84 specific questions to ask the chatbot, resulting from the combination of known failures and social failures, elements from design theories, and syntactic structures. The starting hypothesis to be verified is that a GPT Chat (i.e., a common AI-based chat), properly queried, can provide all the main elements for the automatic compilation of a SFMEA (i.e., to determine the social failures). To do this, the proposed questions were tested in three case studies to extract all the failures and elements that express predefined SFMEA scenarios: a coffee cup provoking gender discrimination, a COVID mask denying a human right, and a thermometer undermining the cultural heritage of a community. The obtained results confirmed the starting hypothesis by showing the strengths and weaknesses of the obtained answers in relation to the following factors: the number and type of inputs (i.e., the failures) provided in the questions; the lexicon used in the question, favoring the use of technical terms derived from design theories and social sustainability taxonomies; the type of the problem. Through this test, the proposed method proved its ability to support the social sustainable design of different products and in different ways. However, a dutiful recommendation instead concerns the tool (i.e., the chatbot) due to its filters that limit some answers in which the designer tries to voluntarily hypothesize failures to explore their social consequences. Full article

Opinion mining refers to the process that helps to identify and to classify users’ emotions and opinions from any source, such as an online review. Thus, opinion mining provides organizations with an insight into their reputation based on previous customers’ opinions regarding their services or products. Automating opinion mining in different languages is still an important topic of interest for scientists, including those using the Arabic language, especially since potential customers mostly do not rate their opinion explicitly. This study proposes an ensemble-based deep learning approach using fastText embeddings and the proposed Arabic emoji and emoticon opinion lexicon to predict user opinion. For testing purposes, the study uses the publicly available Arabic HARD dataset, which includes hotel reviews associated with ratings, starting from one to five. Then, by employing multiple Arabic resources, it experiments with different generated features from the HARD dataset by combining shallow learning with the proposed approach. To the best of our knowledge, this study is the first to create a lexicon that considers emojis and emoticons for its user opinion prediction. Therefore, it is mainly a helpful contribution to the literature related to opinion mining and emojis and emoticons lexicons. Compared to other studies found in the literature related to the five-star rating prediction using the HARD dataset, the accuracy of the prediction using the proposed approach reached an increase of 3.21% using the balanced HARD dataset and an increase of 2.17% using the unbalanced HARD dataset. The proposed work can support a new direction for automating the unrated Arabic opinions in social media, based on five rating levels, to provide potential stakeholders with a precise idea about a service or product quality, instead of spending much time reading other opinions to learn that information. Full article

With the rapid development of science and technology, many new concepts and terms appear, especially in English. Other languages try to express these concepts with words from their vocabulary. In Arabic, there are many ways to find a counterpart for a particularly new concept, such as using an existing word to denote the new concept, derivation, and blending. When these methods fail, the new concepts are phonetically transliterated. Unfortunately, most of the transliterated terms do not conform to the rules of the Arabic language, and many languages, including Arabic, avoid the use of such terms. Some modern linguists call for using the generation strategy to translate new terms into Arabic based on the idea of the meanings of the Arabic letters. Therefore, it is necessary to provide a resource that contains all Arabic roots with a categorization of what is used, what is available for use, and what is rejected according to the phonetic system. This work provides a comprehensive lexicon that contains all possible triliteral roots and determines the status of each root in terms of usage and acceptability. Additionally, it provides a mechanism for giving preference to roots when there is more than one root that indicates the desired meaning. Full article

In recent years, the study of evaluative linguistic expressions has crossed the field of theoretical linguistics and has aroused interest in very different research areas such as artificial intelligence, psychology or cognitive linguistics. The interest in this type of expressions may be due to its relevance in applications such as opinion mining or sentiment analysis. This paper brings together Fuzzy Natural Logic and Fuzzy Property Grammars to approach evaluative expressions. Our contribution includes the marriage of mathematical and linguistic methods for providing a formalism to deal with the linguistic vagueness of evaluative expressions by describing the syntax and semantics of these structures. We contribute to the study of evaluative linguistic expressions by proposing a formal characterization of them where the concepts of semantic prime, borderline evaluative expressions and markedness are defined and where the relation between the semantic constraints of evaluations and their sentiment can be established. A proof-of-concept of how to create a lexicon of evaluative expressions for future computational applications is presented. The results demonstrate that linguistic evaluative expressions are gradient, have sentiment, and that the evaluations work as a relation of hypernym and hyponym, the hypernym being a semantic prime. Our findings provide the basis for building an ontology of evaluative expressions for future computational applications. Full article

Purpose: We aimed to develop a novel interpretable artificial intelligence (AI) model algorithm focusing on automatic detection and classification of various patterns of calcification distribution in mammographic images using a unique graph convolution approach. Materials and methods: Images from 292 patients, which showed calcifications according to the mammographic reports and diagnosed breast cancers, were collected. The calcification distributions were classified as diffuse, segmental, regional, grouped, or linear. Excluded were mammograms with (1) breast cancer with multiple lexicons such as mass, asymmetry, or architectural distortion without calcifications; (2) hidden calcifications that were difficult to mark; or (3) incomplete medical records. Results: A graph-convolutional-network-based model was developed. A total of 581 mammographic images from 292 cases of breast cancer were divided based on the calcification distribution pattern: diffuse (n = 67), regional (n = 115), group (n = 337), linear (n = 8), or segmental (n = 54). The classification performances were measured using metrics including precision, recall, F1 score, accuracy, and multi-class area under the receiver operating characteristic curve. The proposed model achieved a precision of 0.522 ± 0.028, sensitivity of 0.643 ± 0.017, specificity of 0.847 ± 0.009, F1 score of 0.559 ± 0.018, accuracy of 64.325 ± 1.694%, and area under the curve of 0.745 ± 0.030; thus, the method was found to be superior compared to all baseline models. The predicted linear and diffuse classifications were highly similar to the ground truth, and the predicted grouped and regional classifications were also superior compared to baseline models. The prediction results are interpretable using visualization methods to highlight the important calcification nodes in graphs. Conclusions: The proposed deep neural network framework is an AI solution that automatically detects and classifies calcification distribution patterns on mammographic images highly suspected of showing breast cancers. Further study of the AI model in an actual clinical setting and additional data collection will improve its performance. Full article

During the COVID-19 crisis, customers’ preference in having food delivered to their doorstep instead of waiting in a restaurant has propelled the growth of food delivery services (FDSs). With all restaurants going online and bringing FDSs onboard, such as UberEATS, Menulog or Deliveroo, customer reviews on online platforms have become an important source of information about the company’s performance. FDS organisations aim to gather complaints from customer feedback and effectively use the data to determine the areas for improvement to enhance customer satisfaction. This work aimed to review machine learning (ML) and deep learning (DL) models and explainable artificial intelligence (XAI) methods to predict customer sentiments in the FDS domain. A literature review revealed the wide usage of lexicon-based and ML techniques for predicting sentiments through customer reviews in FDS. However, limited studies applying DL techniques were found due to the lack of the model interpretability and explainability of the decisions made. The key findings of this systematic review are as follows: 77% of the models are non-interpretable in nature, and organisations can argue for the explainability and trust in the system. DL models in other domains perform well in terms of accuracy but lack explainability, which can be achieved with XAI implementation. Future research should focus on implementing DL models for sentiment analysis in the FDS domain and incorporating XAI techniques to bring out the explainability of the models. Full article

The huge advancement in Internet web facilities as well as the progress in computing and algorithm development, along with current innovations regarding high-throughput techniques, enable the scientific community to gain access to biological datasets, clinical data and several databases containing billions of pieces of information concerning scientific knowledge. Consequently, during the last decade the system for managing, analyzing, processing and extrapolating information from scientific data has been considerably modified in several fields, including the medical one. As a consequence of the mentioned scenario, scientific vocabulary was enriched by novel lexicons such as machine learning (ML)/deep learning (DL) and overall artificial intelligence (AI). Beyond the terminology, these computational techniques are revolutionizing the scientific research in drug discovery pitch, from the preclinical studies to clinical investigation. Interestingly, between preclinical and clinical research, translational research is benefitting from computer-based approaches, transforming the design and execution of translational research, resulting in breakthroughs for advancing human health. Accordingly, in this review article, we analyze the most advanced applications of AI in translational medicine, providing an up-to-date outlook regarding this emerging field. Full article

Mathematical modeling of language in Artificial Intelligence is of the utmost importance for many research areas and technological applications. Over the last decade, research on text representation has been directed towards the investigation of dense vectors popularly known as word embeddings. In this paper, we propose a cognitive-emotional scoring and representation framework for text based on word embeddings. This representation framework aims to mathematically model the emotional content of words in short free-form text messages, produced by adults in follow-up due to any mental health condition in the outpatient facilities within the Psychiatry Department of Hospital Fundación Jiménez Díaz in Madrid, Spain. Our contribution is a geometrical-topological framework for Sentiment Analysis, that includes a hybrid method that uses a cognitively-based lexicon together with word embeddings to generate graded sentiment scores for words, and a new topological method for clustering dense vector representations in high-dimensional spaces, where points are very sparsely distributed. Our framework is useful in detecting word association topics, emotional scoring patterns, and embedded vectors’ geometrical behavior, which might be useful in understanding language use in this kind of texts. Our proposed scoring system and representation framework might be helpful in studying relations between language and behavior and their use might have a predictive potential to prevent suicide. Full article

Recent work investigating the development of the phonological lexicon, where edges between words represent phonological similarity, have suggested that phonological network growth may be partly driven by a process that favors the acquisition of new words that are phonologically similar to several existing words in the lexicon. To explore this growth mechanism, we conducted a simulation study to examine the properties of networks grown by inverse preferential attachment, where new nodes added to the network tend to connect to existing nodes with fewer edges. Specifically, we analyzed the network structure and degree distributions of artificial networks generated via either preferential attachment, an inverse variant of preferential attachment, or combinations of both network growth mechanisms. The simulations showed that network growth initially driven by preferential attachment followed by inverse preferential attachment led to densely-connected network structures (i.e., smaller diameters and average shortest path lengths), as well as degree distributions that could be characterized by non-power law distributions, analogous to the features of real-world phonological networks. These results provide converging evidence that inverse preferential attachment may play a role in the development of the phonological lexicon and reflect processing costs associated with a mature lexicon structure. Full article