Language Research Articles

Natural Language Processing for Radiation Oncology: Personalizing Treatment Pathways.

Natural language processing (NLP), a technology that translates human language into machine-readable data, is revolutionizing numerous sectors, including cancer care. This review outlines the evolution of NLP and its potential for crafting personalized treatment pathways for cancer patients. Leveraging NLP's ability to transform unstructured medical data into structured learnable formats, researchers can tap into the potential of big data for clinical and research applications. Significant advancements in NLP have spurred interest in developing tools that automate information extraction from clinical text, potentially transforming medical research and clinical practices in radiation oncology. Applications discussed include symptom and toxicity monitoring, identification of social determinants of health, improving patient-physician communication, patient education, and predictive modeling. However, several challenges impede the full realization of NLP's benefits, such as privacy and security concerns, biases in NLP models, and the interpretability and generalizability of these models. Overcoming these challenges necessitates a collaborative effort between computer scientists and the radiation oncology community. This paper serves as a comprehensive guide to understanding the intricacies of NLP algorithms, their performance assessment, past research contributions, and the future of NLP in radiation oncology research and clinics.

Language development, linguistic input, and linguistic racism.

Language development is both remarkable and unremarkable. It is remarkable because children learn the language(s) around them, signed or spoken, without explicit instruction or correction. It is unremarkable because children have done this for thousands of years without worldwide incident or catastrophe. Yet, much research on this organic developmental phenomenon relies on an empirical falsehood: "quality" linguistic input is necessary to facilitate language development. "Quality" is a value judgment, not a structural feature of any human language. I argue selectively legitimizing some linguistic input as "quality" is possible only through mischaracterizing what language is. This falsehood is also linguistic racism because it is based on a deficit perspective of the early linguistic experiences of a subset of children, specifically racialized children. I explore how linguistic racism stalls our collective understanding of language development and promotes an environment of bad science. This article is categorized under: Linguistics > Language Acquisition Psychology > Language Neuroscience > Development.

Artificial Intelligence in Project Management: A Study of The Role of Ai-Powered Chatbots in Project Stakeholder Engagement

Artificial Intelligence (AI) is increasingly becoming a cornerstone in the evolution of project management. Its capabilities extend beyond simple automation, fostering improved decision-making processes and enhancing collaborative efforts. Among the various AI tools available, chatbots stand out as particularly transformative for project management. This study delves into the role of AI-powered chatbots in project stakeholder engagement, a critical aspect of successful project management. Chatbots, powered by sophisticated AI algorithms, can provide continuous support and interaction with project stakeholders. This is particularly vital in managing complex projects where continuous communication and prompt responses can significantly influence project success. The study examines how these AI-driven chatbots facilitate stakeholder engagement, focusing on key benefits such as improved communication, increased stakeholder satisfaction, and better overall project outcomes. Through a detailed analysis, we have identified that chatbots enhance communication by offering stakeholders immediate, personalized responses, thereby reducing response times and improving the efficiency of information exchange. This immediacy and personalization contribute to heightened stakeholder satisfaction, as stakeholders feel their concerns and queries are addressed promptly and effectively. Furthermore, our findings suggest that these improvements in stakeholder engagement directly correlate with enhanced project outcomes, including better adherence to timelines, improved project quality, and increased likelihood of meeting project objectives. However, the deployment of chatbots in project management is not without its challenges. One significant hurdle is the need for advanced natural language processing (NLP) capabilities. Effective chatbots must understand and process complex human language nuances to interact effectively with stakeholders. Another challenge observed is the potential for chatbots to become disruptive or annoying. This can occur when chatbots fail to provide relevant or accurate information, or when their interaction style does not align with stakeholder expectations. In conclusion, AI-powered chatbots hold substantial promise for revolutionizing stakeholder engagement in project management. While they present remarkable benefits in improving communication, stakeholder satisfaction, and project outcomes, there are challenges that need to be addressed. These include enhancing NLP capabilities and fine-tuning the interaction style of chatbots to suit diverse stakeholder groups. With these improvements, AI-powered chatbots could significantly contribute to the success of various projects, marking a new era in project management where AI plays a pivotal role in stakeholder engagement and overall project success.

Conversational facial signals combine into compositional meanings that change the interpretation of speaker intentions

Human language is extremely versatile, combining a limited set of signals in an unlimited number of ways. However, it is unknown whether conversational visual signals feed into the composite utterances with which speakers communicate their intentions. We assessed whether different combinations of visual signals lead to different intent interpretations of the same spoken utterance. Participants viewed a virtual avatar uttering spoken questions while producing single visual signals (i.e., head turn, head tilt, eyebrow raise) or combinations of these signals. After each video, participants classified the communicative intention behind the question. We found that composite utterances combining several visual signals conveyed different meaning compared to utterances accompanied by the single visual signals. However, responses to combinations of signals were more similar to the responses to related, rather than unrelated, individual signals, indicating a consistent influence of the individual visual signals on the whole. This study therefore provides first evidence for compositional, non-additive (i.e., Gestalt-like) perception of multimodal language.

Reprogrammable Non-Linear Circuits Using ReRAM for NN Accelerators

As the massive usage of artificial intelligence techniques spreads in the economy, researchers are exploring new techniques to reduce the energy consumption of Neural Network (NN) applications, especially as the complexity of NNs continues to increase. Using analog Resistive RAM devices to compute matrix-vector multiplication in O (1) time complexity is a promising approach, but it is true that these implementations often fail to cover the diversity of non-linearities required for modern NN applications. In this work, we propose a novel approach where Resistive RAMs themselves can be reprogrammed to compute not only the required matrix multiplications but also the activation functions, Softmax, and pooling layers, reducing energy in complex NNs. This approach offers more versatility for researching novel NN layouts compared to custom logic. Results show that our device outperforms analog and digital field-programmable approaches by up to 8.5× in experiments on real-world human activity recognition and language modeling datasets with convolutional neural network, generative pre-trained Transformer, and long short-term memory models.

Natural language processing analysis of online reviews for small business: extracting insight from small corpora

Receiving and acting on customer input is essential to sustaining and growing any service organization, particularly a small family business whose livelihood depends on strong relationships with its customers. The competitive advantage offered by advanced analytical approaches for supporting decisions is not trivial, and enterprises across virtually all domains of society are investing heavily in this emerging discipline. Natural Language Processing (NLP) is a subset of computer science that employs computational approaches to analyze human language; it is effective at extracting insight from text data but frequently requires large corpora to train its models, in the scale of thousands or millions of documents. This restricts its accessibility to those large enterprises with the capability to capture, store, manage, and analyze such corpora. This research explores a pilot study that applies NLP approaches, specifically topic modeling and large language models (LLM), to assist a small, family-owned business in assessing its strengths and weaknesses based on customer reviews. The relevant corpora of online Facebook, Google Reviews, TripAdvisor, and Yelp reviews is far smaller than ideal, numbering only in the hundreds. Results demonstrate that coherent and actionable insights from big-data approaches are obtainable and that small organizations are not automatically excluded from the benefits of these advanced analytical approaches, with complementary employment of both topic modeling and LLM presenting the greatest potential for similarly-positioned organizations to exploit.

Prediction model of Wordle report based on GA-BP neural network

Natural Language Processing (NLP) plays a vital role in artificial intelligence, enabling machines to understand and generate human language. This paper proposes a novel approach to improve NLP tasks by integrating a Genetic Algorithm-based BP neural network model. The model combines the strengths of genetic algorithms and neural networks. Genetic algorithms offer effective search capabilities, while the BP neural network provides a flexible learning framework. By integrating these techniques, the proposed model aims to overcome limitations of traditional NLP models. It effectively handles the complexity of natural language data, provides efficient training by avoiding local optima, and exhibits enhanced generalization capabilities for unseen data. Extensive experiments on benchmark NLP datasets validate the effectiveness of the proposed model. Results demonstrate its superiority over state-of-the-art NLP models in terms of accuracy, efficiency, and robustness. This paper presents a novel approach to enhance NLP tasks by integrating a genetic algorithm-based BP neural network model. The model shows promising results in various NLP applications and offers advantages over traditional approaches. This research contributes to the advancement of NLP techniques, facilitating more accurate and efficient language processing systems.

Type-shifting in headless relative clauses

Research on the (in)definiteness of bare nouns has developed various proposals regarding which type-shifters exist in human language and which principles are needed to govern their distribution (Carlson 1977; Partee 1987; Chierchia 1998; Dayal 2004; i.a.). At the same time, literature on headless relative clauses (HRCs), primarily focusing on free relatives (FRs) in Indo-European languages, has also developed type-shifting principles (Jacobson 1995; Caponigro 2003, 2004). The type-shifting principles from the FR literature, however, are fundamentally different than those found in proposals for bare nouns. Here, we present case studies from two Mayan languages which diverge from one another in the behavior of bare nouns, and which possess several different kinds of headless relative clauses. We show that “super-free relative clauses” (Caponigro et al. 2021; Caponigro 2022), which lack a wh-word, pattern in ways parallel to bare nouns in the respective languages. We also demonstrate that HRCs headed by a wh-word—i.e., FRs—diverge from bare nouns; they pattern similarly to one another across the languages under investigation, and in ways similar to what has been reported for FRs crosslinguistically. We provide evidence that there is a dedicated FR type-shifter (FR-ι) used as a post-syntactic mechanism to repair a type-mismatch at the CP level, building on work by Caponigro (2004). Our novel contribution is that this type-shifter is available regardless of the presence or absence of other type-shifters in a language. This paper adds new data to our understanding of the range and applicability of different definiteness-related type-shifters as well as captures certain typological tendencies regarding HRCs.

SRBerta—A Transformer Language Model for Serbian Cyrillic Legal Texts

Language is a unique ability of human beings. Although relatively simple for humans, the ability to understand human language is a highly complex task for machines. For a machine to learn a particular language, it must understand not only the words and rules used in a particular language, but also the context of sentences and the meaning that words take on in a particular context. In the experimental development we present in this paper, the goal was the development of the language model SRBerta—a language model designed to understand the formal language of Serbian legal documents. SRBerta is the first of its kind since it has been trained using Cyrillic legal texts contained within a dataset created specifically for this purpose. The main goal of SRBerta network development was to understand the formal language of Serbian legislation. The training process was carried out using minimal resources (single NVIDIA Quadro RTX 5000 GPU) and performed in two phases—base model training and fine-tuning. We will present the structure of the model, the structure of the training datasets, the training process, and the evaluation results. Further, we will explain the accuracy metric used in our case and demonstrate that SRBerta achieves a high level of accuracy for the task of masked language modeling in Serbian Cyrillic legal texts. Finally, SRBerta model and training datasets are publicly available for scientific and commercial purposes.

GPT-Driven Source-to-Source Transformation for Generating Compilable Parallel CUDA Code for Nussinov’s Algorithm

Designing automatic optimizing compilers is an advanced engineering process requiring a great deal of expertise, programming, testing, and experimentation. Maintaining the approach and adapting it to evolving libraries and environments is a time-consuming effort. In recent years, OpenAI has presented the GPT model, which is designed for many fields like computer science, image processing, linguistics, and medicine. It also supports automatic programming and translation between programming languages, as well as human languages. This article will verify the usability of the commonly known LLM model, GPT, for the non-trivial NPDP Nussinov’s parallel algorithm code within the OpenMP standard to create a parallel equivalent of CUDA for NVIDIA graphics cards. The goal of this approach is to avoid creating any post-processing scripts and writing any lines of target code. To validate the output code, we compare the resulting arrays with the ones calculated by the optimized code for the CPU generated employing the polyhedral compilers. Finally, the code will be checked for scalability and performance. We will concentrate on assessing the capabilities of GPT, highlighting common challenges that can be refined during future learning processes. This will enhance code generation for various platforms by leveraging the outcomes from polyhedral optimizers.

Applications of Advanced Natural Language Processing for Clinical Pharmacology.

Natural language processing (NLP) is a branch of artificial intelligence, which combines computational linguistics, machine learning, and deep learning models to process human language. Although there is a surge in NLP usage across various industries in recent years, NLP has not been widely evaluated and utilized to support drug development. To demonstrate how advanced NLP can expedite the extraction and analyses of information to help address clinical pharmacology questions, inform clinical trial designs, and support drug development, three use cases are described in this article: (1) dose optimization strategy in oncology, (2) common covariates on pharmacokinetic (PK) parameters in oncology, and (3) physiologically-based PK (PBPK) analyses for regulatory review and product label. The NLP workflow includes (1) preparation of source files, (2) NLP model building, and (3) automation of data extraction. The Clinical Pharmacology and Biopharmaceutics Summary Basis of Approval (SBA) documents, US package inserts (USPI), and approval letters from the US Food and Drug Administration (FDA) were used as our source data. As demonstrated in the three example use cases, advanced NLP can expedite the extraction and analyses of large amounts of information from regulatory review documents to help address important clinical pharmacology questions. Although this has not been adopted widely, integrating advanced NLP into the clinical pharmacology workflow can increase efficiency in extracting impactful information to advance drug development.

Exploring data-driven chemical SMILES tokenization approaches to identify key protein-ligand binding moieties.

Machine learning models have found numerous successful applications in computational drug discovery. A large body of these models represents molecules as sequences since molecular sequences are easily available, simple, and informative. The sequence-based models often segment molecular sequences into pieces called chemical words, analogous to the words that make up sentences in human languages, and then apply advanced natural language processing techniques for tasks such as de novo drug design, property prediction, and binding affinity prediction. However, the chemical characteristics and significance of these building blocks, chemical words, remain unexplored. To address this gap, we employ data-driven SMILES tokenization techniques such as Byte Pair Encoding, WordPiece, and Unigram to identify chemical words and compare the resulting vocabularies. To understand the chemical significance of these words, we build a language-inspired pipeline that treats high affinity ligands of protein targets as documents and selects key chemical words making up those ligands based on tf-idf weighting. The experiments on multiple protein-ligand affinity datasets show that despite differences in words, lengths, and validity among the vocabularies generated by different subword tokenization algorithms, the identified key chemical words exhibit similarity. Further, we conduct case studies on a number of target to analyze the impact of key chemical words on binding. We find that these key chemical words are specific to protein targets and correspond to known pharmacophores and functional groups. Our approach elucidates chemical properties of the words identified by machine learning models and can be used in drug discovery studies to determine significant chemical moieties.

Potential applications of ChatGPT in obstetrics and gynecology in Korea: a review article.

The use of chatbot technology, particularly chat generative pre-trained transformer (ChatGPT) with an impressive 175 billion parameters, has garnered significant attention across various domains, including Obstetrics and Gynecology (OBGYN). This comprehensive review delves into the transformative potential of chatbots with a special focus on ChatGPT as a leading artificial intelligence (AI) technology. Moreover, ChatGPT harnesses the power of deep learning algorithms to generate responses that closely mimic human language, opening up myriad applications in medicine, research, and education. In the field of medicine, ChatGPT plays a pivotal role in diagnosis, treatment, and personalized patient education. Notably, the technology has demonstrated remarkable capabilities, surpassing human performance in OBGYN examinations, and delivering highly accurate diagnoses. However, challenges remain, including the need to verify the accuracy of the information and address the ethical considerations and limitations. In the wide scope of chatbot technology, AI systems play a vital role in healthcare processes, including documentation, diagnosis, research, and education. Although promising, the limitations and occasional inaccuracies require validation by healthcare professionals. This review also examined global chatbot adoption in healthcare, emphasizing the need for user awareness to ensure patient safety. Chatbot technology holds great promise in OBGYN and medicine, offering innovative solutions while necessitating responsible integration to ensure patient care and safety.

Hubungan antara Islam dan Kebudayaan Jawa

The discussion on the relationship between Islam and Javanese culture in Indonesia is intriguing. Culture is defined as a complex system that encompasses values, norms, beliefs, language, art, customs, institutions, technology, and creative products of humans. The focus od the research is on Javanese culture, which is prominent in Indonesia. The connection between Islam and Javanese culture is reflected in a long and diverse history, influenced by various ethnic groups, religions, and cultures united within one nation. This research adopts a literature review method, analyzing information from various literary sources such as books, research journals, and scholarly works. The conclusion indicates that the relationship between Islam and Javanese culture is inseparable, despite some contradictions. A thorough study of Islamic teachings is necessary to ensure proper integration of religious values and cultural traditions in the daily lives of the Javanese community.

Understanding through the Numbers: Number Systems, Their Evolution, and Their Perception among Kula People from Alor Island, Southeastern Indonesia

This paper aims at documenting and reconstructing the linguistic processes generating and substantiating the use of number systems, numbers in general, elementary arithmetic, and the related concepts and notions among the Kula people from Alor Island, Southeastern Indonesia. The Kula is a Papuan population from the Alor–Pantar Archipelago (Timor area). The name of their language, Kula (or Kola), corresponds to the ethnonym. The language is, currently, endangered and not completely documented. At the level of linguistic features, numeral systems and the terms for numerals from Eastern Alor exhibit, to some extent, unique characteristics, if compared to other languages spoken in other sectors of the island. Therefore, the Kula numbering system is not only significant at the lexicological and lexicographic level, but also represents the essential role of cognitive strategies (e.g., the choice of the base for the numbering systems and the visual representation of counting with the aid of actual ‘objects’, like hands and fingers) in the coinage of numerical terms among the local speakers. Indeed, the development of numeral systems reflects the evolution of human language and the ability of humans to construct abstract numerical concepts. The way numerals are encoded and expressed in a language can impact the patterns according to which numerical notions are conceptualized and understood. Different numeral systems can indicate variations in cognitive processes involving notions of quantities and measurements. Therefore, the structure and characteristics of a numeral system may affect how numeral concepts are mentally represented and developed. This paper focuses on the number system of the Kula people and the lexical units used by the local speakers to indicate (and to explain) the numbers, with the related concepts, notions, and symbolism. The investigation delves into the degrees of abstraction of the Kula numeral system and tries to ascertain its origins and reconstruct it. Moreover, the article applies to the analysis a comparative approach, which takes into account several Papuan and Austronesian languages from Alor Island and Eastern Timor, with the dual aim of investigating, at a preliminary level, a possible common evolution and/or divergent naming processes in local numbering systems and their historical–linguistic and etymological origins.

Jürgen Habermas’s Translation of the Human Being as Created in the Image of God: Perspectives from Joseph Ratzinger and Alasdair MacIntyre

This paper considers Habermas’s translation proviso, which requires religious concepts to be translated into secular language when in the public sphere. Translation, for Habermas, protects the state from religious interference and elicits essential aspects of pre-rational thought—that is, religious and metaphysical thought, which post-metaphysics cannot generate for itself, e.g., social solidarity. The task undertaken by Habermas’s translation proviso is illustrated through his own work of translation: that of the translation of the biblical image of humanity as created in the image of God into the identical dignity of each human being. To provide context to and to highlight the difficulties involved in Habermas’s translation proviso, consideration is given to the thought of Joseph Ratzinger/Pope Benedict XVI and Alasdair MacIntyre on these themes. What is demonstrated is that Habermas’s translation is, in essence, assimilation and re-appropriation. In practice, it manifests itself as the truncation of Christian metaphysics, in which the divine Logos is replaced by or collapsed into the logos of intersubjective human language. The relational image of humanity as a creature distinct from the Creator, in which human reason is analogous to divine reason, is erased, leaving autonomous human beings, from which human reason emerges out of the discursive communication of the logos of intersubjective human language. The conclusion is that the translation proviso fails in its objective. An alternative to Habermas’s translation proviso, the presupposition proviso, is presented as a more apt approach to addressing the underlying issues involved: facilitating human flourishing in an orderly, free, and just society.

Availability of ChatGPT to provide medical information for patients with kidney cancer

ChatGPT is an advanced natural language processing technology that closely resembles human language. We evaluated whether ChatGPT could help patients understand kidney cancer and replace consultations with urologists. Two urologists developed ten questions commonly asked by patients with kidney cancer. The answers to these questions were produced using ChatGPT. The five-dimension SERVQUAL model was used to assess the service quality of ChatGPT. The survey was distributed to 103 urologists via email, and twenty-four urological oncologists specializing in kidney cancer were included as experts with more than 20 kidney cancer cases in clinic per month. All respondents were physicians. We received 24 responses to the email survey (response rate: 23.3%). The appropriateness rate for all ten answers exceeded 60%. The answer to Q2 received the highest agreement (91.7%, etiology of kidney cancer), whereas the answer to Q8 had the lowest (62.5%, comparison with other cancers). The experts gave low assessment ratings (44.4% vs. 93.3%, p = 0.028) in the SERVQUAL assurance (certainty of total answers) dimension. Positive scores for the overall understandability of ChatGPT answers were assigned by 54.2% of responders, and 70.8% said that ChatGPT could not replace explanations provided by urologists. Our findings affirm that although ChatGPT answers to kidney cancer questions are generally accessible, they should not supplant the counseling of a urologist.

Research on Human Language and Cognition in Brazil Based on Stimulus Equivalence and Relational Frame Theory

Research on Human Language and Cognition in Brazil Based on Stimulus Equivalence and Relational Frame Theory

Bodies at play: the role of intercorporeality and bodily affordances in coordinating social play in chimpanzees in the wild.

The comparative approach is a crucial method to gain a better understanding of the behavior of living human and nonhuman animals to then draw informed inferences about the behavior of extinct ancestors. One focus has been on disentangling the puzzle of language evolution. Traditionally, studies have predominantly focused on intentionally produced signals in communicative interactions. However, in collaborative and highly dynamic interactions such as play, underlying intentionality is difficult to assess and often interactions are negotiated via body movements rather than signals. This "lack" of signals has led to this dynamic context being widely ignored in comparative studies. The aim of this paper is threefold: First, we will show how comparative research into communication can benefit from taking the intentionality-agnostic standpoint used in conversation analysis. Second, we will introduce the concepts of 'intercorporeality' and 'bodily affordance', and show how they can be applied to the analysis of communicative interactions of nonhuman animals. Third, we will use these concepts to investigate how chimpanzees (Pan troglodytes) initiate, end, and maintain 'contact social play'. Our results showed that bodily affordances are able to capture elements of interactions that more traditional approaches failed to describe. Participants made use of bodily affordances to achieve coordinated engagement in contact social play. Additionally, these interactions could display a sequential organization by which one 'move' by a chimpanzee was responded to with an aligning 'move', which allowed for the co-construction of the activity underway. Overall, the present approach innovates on three fronts: First, it allows for the analysis of interactions that are often ignored because they do not fulfil criteria of intentionality, and/or consist of purely body movements. Second, adopting concepts from research on human interaction enables a better comparison of communicative interactions in other animal species without a too narrow focus on intentional signaling only. Third, adopting a stance from interaction research that highlights how practical action can also be communicative, our results show that chimpanzees can communicate through their embodied actions as well as through signaling. With this first step, we hope to inspire new research into dynamic day-to-day interactions involving both "traditional" signals and embodied actions, which, in turn, can provide insights into evolutionary precursors of human language.

Surfing the OCEAN: The machine learning psycholexical approach 2.0 to detect personality traits in texts.

We aimed to develop a machine learning model to infer OCEAN traits from text. The psycholexical approach allows retrieving information about personality traits from human language. However, it has rarely been applied because of methodological and practical issues that current computational advancements could overcome. Classical taxonomies and a large Yelp corpus were leveraged to learn an embedding for each personality trait. These embeddings were used to train a feedforward neural network for predicting trait values. Their generalization performances have been evaluated through two external validation studies involving experts (N = 11) and laypeople (N = 100) in a discrimination task about the best markers of each trait and polarity. Intrinsic validation of the model yielded excellent results, with R2 values greater than 0.78. The validation studies showed a high proportion of matches between participants' choices and model predictions, confirming its efficacy in identifying new terms related to the OCEAN traits. The best performance was observed for agreeableness and extraversion, especially for their positive polarities. The model was less efficient in identifying the negative polarity of openness and conscientiousness. This innovative methodology can be considered a "psycholexical approach 2.0," contributing to research in personality and its practical applications in many fields.