This technical article explores the evolution and current state of Natural Language Processing (NLP), focusing on its fundamental components, sentiment analysis capabilities, language generation techniques, and implementation considerations. The article examines the transformation of NLP through transformer-based architectures, discussing advancements in text preprocessing, tokenization methods, and named entity recognition. It analyzes the progression of sentiment analysis from basic lexicon-based approaches to sophisticated neural architectures, highlighting improvements in contextual understanding and emotional context detection. The article also investigates (...) modern language generation systems, their architectural innovations, and practical applications. Additionally, it addresses critical implementation considerations, including computational requirements, data quality concerns, and ethical implications, providing insights into the deployment challenges and solutions in real-world NLP applications. (shrink)

In the digital age, chatbots have emerged as essential tools for automating communication and improving user experiences across various sectors. This paper presents a Chatbot Assistant System powered by Natural Language Processing (NLP) to provide intelligent, context-aware, and real-time responses to user queries. The system incorporates NLP techniques, such as text preprocessing, intent recognition, and entity extraction, to facilitate effective interactions. We explore the architecture, working principles, and applications of the system, along with its performance evaluation in (...) different domains. (shrink)

Knowledge is power. Knowledge about human psychology is increasingly being produced using natural language processing (NLP) and related techniques. The power that accompanies and harnesses this knowledge should be subject to ethical controls and oversight. In this chapter, we address the ethical pitfalls that are likely to be encountered in the context of such research. These pitfalls occur at various stages of the NLP pipeline, including data acquisition, enrichment, analysis, storage, and sharing. We also address secondary uses (...) of the results and tools developed through psychometric NLP, such as profit-driven targeted advertising, political campaigns, and domestic and international psyops. Along the way, we reflect on potential ethical guidelines and considerations that may help researchers navigate these pitfalls. (shrink)

Despite its centrality in the philosophy of cognitive science, there has been little prior philosophical work engaging with the notion of representation in contemporary NLP practice. This paper attempts to fill that lacuna: drawing on ideas from cognitive science, I introduce a framework for evaluating the representational claims made about components of neural NLP models, proposing three criteria with which to evaluate whether a component of a model represents a property and operationalising these criteria using probing classifiers, a popular analysis (...) technique in NLP (and deep learning more broadly). The project of operationalising a philosophically-informed notion of representation should be of interest to both philosophers of science and NLP practitioners. It affords philosophers a novel testing-ground for claims about the nature of representation, and helps NLPers organise the large literature on probing experiments, suggesting novel avenues for empirical research. (shrink)

Text classification is one of the most important task in natural language processing, In this research, we carried out several experimental research on three (3) of the most popular Text classification NLP classifier in Convolutional Neural Network (CNN), Multinomial Naive Bayes (MNB), and Support Vector Machine (SVN). In the presence of enough training data, Deep Learning CNN work best in all parameters for evaluation with 77% accuracy, followed by SVM with accuracy of 76%, and multinomial Bayes with (...) least performance of 69% accuracy. CNN has the best performance in the presence of large enough training dataset because of the presence of filter/ kernels which help to indentify patterns in text data regardless of their position in the sentence. We repeated the training again with just one-third of our data, at this point SVM comes with the best performance, the performance of CNN noticeably drops but still better than Multinomial Naive Bayes, the reason why SVM works best when we reduce the training data was because of its ability to look for a hyper-plane that creates a boundary between different classes of data so as to properly classify them, so we believed that getting the hyper-plane was more efficient when we reduce the dataset, hence reason for the good performance. Multinomial Naive Bayes have the least performance which we attributed to its assumption of independency between the features which sometimes does not hold true. We concluded that availability of data should be an important factor when choosing classifier for Natural Language Processing Text Classification task. CNN should be use in the presence of enough dataset, and SVM should be use when data is not enough. Multinomial Naive Bayes must not be trusted with state of the art NLP task due to its assumption of independency between the features . (shrink)

The rise of social media has created a new platform for communication and interaction, but it has also facilitated the spread of harmful behaviors such as cyberbullying. Detecting and mitigating cyberbullying on social media platforms is a critical challenge that requires advanced technological solutions. This paper presents a novel approach to cyberbullying detection using a combination of supervised machine learning (ML) and natural language processing (NLP) techniques, enhanced by optimization algorithms. The proposed system is designed to identify (...) and classify cyberbullying behavior in real-time, analyzing textual data from social media posts to detect harmful content. The model is trained on a large dataset of labeled instances of bullying and non-bullying content, using supervised ML algorithms such as Support Vector Machines (SVM), Decision Trees, and Random Forest. NLP techniques, including sentiment analysis, keyword extraction, and text vectorization, are employed to preprocess and transform the data into a format suitable for machine learning. T. (shrink)

Customer feedback and reviews are rich sources of information that reflect the sentiments and experiences of consumers, especially in the financial sector. Mining customer sentiments from these textual data sources provides valuable insights for improving services, identifying emerging issues, and predicting customer satisfaction. This paper proposes a novel approach to mining customer sentiments from financial feedback and reviews, leveraging advanced natural language processing (NLP) techniques, sentiment analysis algorithms, and machine learning models. We discuss methods for preprocessing financial (...) text data, feature extraction, sentiment classification, and visualization of sentiment trends. Additionally, we explore the application of sentiment analysis in understanding customer behaviour, improving financial product offerings, and enhancing customer support strategies. Experimental results demonstrate the effectiveness of the proposed methods in accurately predicting customer sentiments and generating actionable insights. This paper also suggests directions for future research, focusing on real-time sentiment monitoring and integration with personalized financial services. (shrink)

Customer feedback and reviews are rich sources of information that reflect the sentiments and experiences of consumers, especially in the financial sector. Mining customer sentiments from these textual data sources provides valuable insights for improving services, identifying emerging issues, and predicting customer satisfaction. This paper proposes a novel approach to mining customer sentiments from financial feedback and reviews, leveraging advanced natural language processing (NLP) techniques, sentiment analysis algorithms, and machine learning models. We discuss methods for preprocessing financial (...) text data, feature extraction, sentiment classification, and visualization of sentiment trends. Additionally, we explore the application of sentiment analysis in understanding customer behaviour, improving financial product offerings, and enhancing customer support strategies. Experimental results demonstrate the effectiveness of the proposed methods in accurately predicting customer sentiments and generating actionable insights. This paper also suggests directions for future research, focusing on real-time sentiment monitoring and integration with personalized financial services. (shrink)

The rise of social media has created a new platform for communication and interaction, but it has also facilitated the spread of harmful behaviors such as cyberbullying. Detecting and mitigating cyberbullying on social media platforms is a critical challenge that requires advanced technological solutions. This paper presents a novel approach to cyberbullying detection using a combination of supervised machine learning (ML) and natural language processing (NLP) techniques, enhanced by optimization algorithms. The proposed system is designed to identify (...) and classify cyberbullying behavior in real-time, analyzing textual data from social media posts to detect harmful content. The model is trained on a large dataset of labeled instances of bullying and non-bullying content, using supervised ML algorithms such as Support Vector Machines (SVM), Decision Trees, and Random Forest. NLP techniques, including sentiment analysis, keyword extraction, and text vectorization, are employed to preprocess and transform the data into a format suitable for machine learning. To optimize the performance of the detection model, techniques such as Grid Search, Genetic Algorithms, and Particle Swarm Optimization are used to fine-tune hyperparameters, resulting in improved accuracy and reduced false positives. The system's effectiveness is validated through experiments conducted on various social media platforms, demonstrating its potential to detect cyberbullying with high precision. Future work will focus on enhancing the model's adaptability to emerging slang and evolving language patterns in social media. (shrink)

Text classification is one of the most important task in natural language processing, In this research, we carried out several experimental research on three (3) of the most popular Text classification NLP classifier in Convolutional Neural Network (CNN), Multinomial Naive Bayes (MNB), and Support Vector Machine (SVN). In the presence of enough training data, Deep Learning CNN work best in all parameters for evaluation with 77% accuracy, followed by SVM with accuracy of 76%, and multinomial Bayes with (...) least performance of 69% accuracy. CNN has the best performance in the presence of large enough training dataset because of the presence of filter/ kernels which help to indentify patterns in text data regardless of their position in the sentence. -/- We repeated the training again with just one-third of our data, at this point SVM comes with the best performance, the performance of CNN noticeably drops but still better than Multinomial Naive Bayes, the reason why SVM works best when we reduce the training data was because of its ability to look for a hyper-plane that creates a boundary between different classes of data so as to properly classify them, so we believed that getting the hyper-plane was more efficient when we reduce the dataset, hence reason for the good performance. Multinomial Naive Bayes have the least performance which we attributed to its assumption of independency between the features which sometimes does not hold true. We concluded that availability of data should be an important factor when choosing classifier for Natural Language Processing Text Classification task. CNN should be use in the presence of enough dataset, and SVM should be use when data is not enough. Multinomial Naive Bayes must not be trusted with state of the art NLP task due to its assumption of independency between the features. (shrink)

Artificial Intelligence (AI) technologies are now widely used in a variety of fields to aid with knowledge acquisition and decision-making. Health information systems, in particular, can gain the most from AI advantages. Recently, symptoms-based illness prediction research and manufacturing have grown in popularity in the healthcare business. Several scholars and organisations have expressed an interest in applying contemporary computational tools to analyse and create novel approaches for rapidly and accurately predicting illnesses. In this study, we present a paradigm for assessing (...) the efficacy of combining Machine Learning (ML) and Natural Language Processing (NLP) technologies in a disease prediction system. We scraped a disease-symptom dataset with NLP characteristics from one of the UK's most trusted National Health Service (NHS) websites as an example. In addition, we will thoroughly examine our data using symptom frequency, similarity, and clustering analysis. As a consequence, we can observe that the forecast has a high efficiency rate, but there are still some challenges to work out. (shrink)

Emotions are a central aspect of communication. Consequently, emotion analysis (EA) is a rapidly growing field in natural language processing (NLP). However, there is no consensus on scope, direction, or methods. In this paper, we conduct a thorough review of 154 relevant NLP publications from the last decade. Based on this review, we address four different questions: (1) How are EA tasks defined in NLP? (2) What are the most prominent emotion frameworks and which emotions are modeled? (...) (3) Is the subjectivity of emotions considered in terms of demographics and cultural factors? and (4) What are the primary NLP applications for EA? We take stock of trends in EA and tasks, emotion frameworks used, existing datasets, methods, and applications. We then discuss four lacunae: (1) the absence of demographic and cultural aspects does not account for the variation in how emotions are perceived, but instead assumes they are universally experienced in the same manner; (2) the poor fit of emotion categories from the two main emotion theories to the task; (3) the lack of standardized EA terminology hinders gap identification, comparison, and future goals; and (4) the absence of interdisciplinary research isolates EA from insights in other fields. Our work will enable more focused research into EA and a more holistic approach to modeling emotions in NLP. (shrink)

Phishing, a prevalent form of social engineering attack, continues to threaten enterprises by exploiting human vulnerabilities and targeting sensitive information. With the increasing sophistication of phishing schemes, traditional detection methods often fall short in identifying and mitigating these threats. As attackers employ advanced techniques, such as highly personalized spear-phishing emails and malicious links, enterprises require innovative solutions to safeguard their digital ecosystems. This research explores the application of artificial intelligence (AI) in enhancing phishing detection and response, with a specific focus (...) on addressing advanced social engineering tactics. AIpowered phishing detection systems utilize machine learning (ML) and natural language processing (NLP) to analyze communication patterns, detect anomalies, and identify deceptive content in real-time. Unlike conventional rule-based systems, AI approaches adapt dynamically to evolving attack vectors, offering a robust defense against both known and emerging threats. These systems can scrutinize email headers, content, URLs, and attachments to uncover subtle indicators of malicious intent. By leveraging historical datasets and predictive analytics, AI models identify patterns associated with phishing, such as unusual sender behavior, suspicious linguistic cues, and mismatched domain information. This study examines the critical components of AI-driven phishing detection, including feature extraction, algorithm selection, and system deployment within enterprise environments. It highlights the role of NLP in understanding the contextual and linguistic subtleties of phishing messages and how advanced ML models, such as decision trees, support vector machines, and deep neural networks, enhance detection accuracy. -/- Additionally, the research delves into the challenges of implementing AI in phishing detection, such as ensuring data privacy, mitigating adversarial attacks on AI models, and fostering trust through explainable AI. Real-world case studies are analyzed to demonstrate the practical effectiveness of AI-powered phishing detection systems. These examples showcase significant reductions in phishing incidents, improved detection times, and the ability to thwart highly targeted attacks, such as business email compromise (BEC) schemes. The study also emphasizes the importance of integrating AI solutions with existing cybersecurity frameworks and highlights the role of human oversight in complementing AI capabilities. While AI offers promising advancements, its success depends on robust training datasets, continuous model optimization, and alignment with enterprise security goals. The paper concludes by presenting best practices for deploying AI-based phishing detection, including regular model updates, employee awareness programs, and multi-layered security strategies. By leveraging AI, enterprises can proactively defend against advanced social engineering attacks, safeguarding their systems, data, and stakeholders from evolving cyber threats. This research contributes to the growing field of AI applications in cybersecurity, offering actionable insights for organizations seeking to enhance their resilience against sophisticated phishing campaigns. (shrink)

In a country as linguistically diverse as India, the demand for localized video content is rapidly increasing. Dubbing English videos into Indian languages, such as Hindi, Tamil, Telugu, and Bengali, is critical to making content accessible to wider audiences. This paper presents a software solution that automates the dubbing process using a combination of technologies like Automatic Speech Recognition (ASR), Machine Translation (MT), andText-to-Speech (TTS). The process begins by converting English audio into text using ASR, followed by translation into the (...) target Indian language through advanced Natural Language Processing (NLP) models. The translated text is then synthesized into natural-sounding speech using TTS engines like Google Text-to-Speech (gTTS). Finally, the dubbed audio is synchronized with the original video to ensure proper lip-sync and timing. This automated approach streamlines the dubbing process, reducing time and cost while maintaining translation accuracy, voice quality, and synchronization. The software offers an efficient solution for content creators, businesses, and educators looking to expand their reach across India’s multilingual population. (shrink)

The main objective of our project is to construct a fully functional voice-based home automation system that uses the Internet of Things and Natural Language Processing. The home automation system is user-friendly to smartphones and laptops. A set of relays is used to connect the Node MCU to homes under controlled appliances. The user sends a command through the speech to the mobile devices, which interprets the message and sends the appropriate command to the specific appliance. The (...) voice command given by the user is interpreted by the mobile device using Natural Language Processing. The system utilizes a node microcontroller unit (Node MCU) as a Wi-Fi-based gateway to connect to the Google Assistant. It determines operation must be completed by which the appliance to fulfill the user’s request. The user increased mobile capabilities we will be using a smart phone in this project. A smartphone application as nearly all of the data transferred can be processed by cloud. The user increased mobile capabilities we will be using a smart phone in this project. The IoT -based system for home automation can easily and efficiently control appliances. The Node MCU is interfaced with the appliances and programmed in a manner that they respond to mobile inputs. (shrink)

This research paper presents an in-depth analysis of advanced artificial intelligence (AI) algorithms designed to automate data preprocessing in the healthcare sector. The automation of data preprocessing is crucial due to the overwhelming volume, diversity, and complexity of healthcare data, which includes medical records, diagnostic imaging, sensor data from medical devices, genomic data, and other heterogeneous sources. These datasets often exhibit various inconsistencies such as missing values, noise, outliers, and redundant or irrelevant information that necessitate extensive preprocessing before being analyzed (...) by machine learning or statistical models. Traditional data preprocessing methods, which are largely manual and time-consuming, can result in errors that affect the quality of the data and, subsequently, the performance of predictive and diagnostic models. Thus, there is a growing need for intelligent, automated systems that can enhance data quality, streamline the preprocessing pipeline, and reduce the time and effort required by healthcare professionals and data scientists. The study begins by outlining the specific challenges associated with healthcare data, including its high dimensionality, incompleteness, and variability across different data sources and formats. These issues not only complicate the preprocessing stage but also hinder the ability to develop robust models capable of making accurate predictions or diagnoses. The paper then explores how AI algorithms—particularly those based on machine learning (ML), deep learning (DL), and reinforcement learning (RL)—can automate key data preprocessing tasks such as data cleaning, feature selection, normalization, and transformation. These algorithms are designed to identify patterns in data, detect anomalies, and automatically apply corrections or transformations based on predefined rules or learned behaviors, thereby minimizing human intervention. The paper also delves into specific AI techniques that have been successfully applied to healthcare data preprocessing. For instance, supervised learning models, such as decision trees and support vector machines (SVMs), have been utilized to perform imputation of missing data by predicting the most likely values based on the available information. Similarly, unsupervised learning methods, such as clustering algorithms, have been employed to group similar data points and remove outliers that could distort the performance of analytical models. Moreover, deep learning techniques, particularly autoencoders and generative adversarial networks (GANs), have demonstrated remarkable effectiveness in transforming high-dimensional medical data into lower-dimensional representations, enabling more efficient and accurate model training. In addition to the discussion of these algorithms, the paper emphasizes the role of natural language processing (NLP) in automating the preprocessing of unstructured healthcare data, such as clinical notes and diagnostic reports. NLP techniques, including named entity recognition (NER) and word embeddings, are instrumental in extracting relevant information from unstructured text, standardizing terminologies, and converting textual data into structured formats suitable for downstream analysis. Furthermore, AI-based feature selection algorithms are explored, which aim to identify the most relevant features in the dataset, thereby reducing its dimensionality and improving the computational efficiency of predictive models. The study goes on to highlight the significant reduction in processing time achieved by AIdriven automation of preprocessing tasks. In conventional settings, data preprocessing accounts for a substantial portion of the time spent on building healthcare models, often requiring expert intervention to manually inspect and clean the data. By employing AI algorithms, not only can this process be expedited, but the accuracy of the resulting data is also enhanced, which translates into better model performance. The paper provides a detailed comparative analysis of manual preprocessing methods versus automated AI-driven approaches, demonstrating the substantial time savings and improvements in data quality brought about by automation. (shrink)

The rise of social media has created a new platform for communication and interaction, but it has also facilitated the spread of harmful behaviors such as cyberbullying. Detecting and mitigating cyberbullying on social media platforms is a critical challenge that requires advanced technological solutions. This paper presents a novel approach to cyberbullying detection using a combination of supervised machine learning (ML) and natural language processing (NLP) techniques, enhanced by optimization algorithms. The proposed system is designed to identify (...) and classify cyberbullying behavior in real-time, analyzing textual data from social media posts to detect harmful content. The model is trained on a large dataset of labeled instances of bullying and non-bullying content, using supervised ML algorithms such as Support Vector Machines (SVM), Decision Trees, and Random Forest. NLP techniques, including sentiment analysis, keyword extraction, and text vectorization, are employed to preprocess and transform the data into a format suitable for machine learning. To optimize the performance of the detection model, techniques such as Grid Search, Genetic Algorithms, and Particle Swarm Optimization are used to fine-tune hyperparameters, resulting in improved accuracy and reduced false positives. The system's effectiveness is validated through experiments conducted on various social media platforms, demonstrating its potential to detect cyberbullying with high precision. Future work will focus on enhancing the model's adaptability to emerging slang and evolving language patterns in social media. Key words: Cyberbullying Detection, Social Media, Supervised Machine Learning, Natural Language Processing (NLP), Optimization Techniques. (shrink)

Ayurvedic treatment, a traditional system of medicine, is valued for its individualized, and natural approach to health. Ayurvedic treatment encompasses a wide range of systems and practices designed to promote health and treat illness through holistic approaches. Existing solutions for Ayurvedic treatments include Ayur Times, Med9, and Jark Pharma. It includes diagnostic methods, various treatment modalities, and lifestyle recommendations. There are issues in the existing system, as in, the system lacks personalized recommendations, a user-friendly interface, and tailored filtering options. (...) The existing system does not fully address the need for a holistic, user-centric approach combining detailed information with personalized recommendations. This project aims to develop a website to assist Ayurvedic practitioners and students in finding the most suitable herbal treatments. Users will input symptoms, health conditions, and patient details such as allergies and dietary restrictions, which will be used to search a comprehensive database of Ayurvedic texts for personalized recommendations. (shrink)

The reproducibility crisis in social sciences has revealed significant weaknesses in conventional research practices, including selective publication, questionable statistical methods, and opaque peer review processes. This paper introduces Granular Interaction Thinking Theory (GITT) as a novel framework for understanding the plausibility of scientific findings, conceptualizing knowledge validation as a structured entropy-reduction process. Within this framework, open science practices—such as open data, open review, and open dialogue—initially increase informational entropy by exposing inconsistencies. However, through iterative refinement, they ultimately enhance the robustness (...) and plausibility of scientific knowledge. To systematically assess a study’s plausibility, we propose the Scientific Plausibility Index (SPI)—an entropy-based metric that integrates data and method transparency, replication success rates, and community engagement. Additionally, leveraging artificial intelligence (AI) and natural language processing (NLP), a dynamic plausibility-tracking system could be developed to detect unreliable claims early and accelerate scientific self-correction. This shift—from closed, one-time assessments to continuous, community-driven evaluation models—ensures that scientific knowledge remains rigorously tested and updated. We advocate for shifting from closed, one-time assessments to continuous, community-driven evaluation models, ensuring that scientific knowledge remains rigorously tested and updated. Ultimately, aligning incentives with entropy-reducing mechanisms—such as rewarding replication efforts and fostering open discourse—can cultivate a research culture that prioritizes robustness, transparency, and cumulative knowledge-building. (shrink)

Indian Sign Language (ISL) users face challenges communicating effectively with English speakers, primarily due to a lack of accessible, real-time translation tools. Existing systems often struggle with context and natural flow, leading to misunderstandings and reduced usability. This project proposes an AI-powered system designed to overcome these limitations, incorporating speech recognition and natural language processing (NLP) to facilitate seamless interaction between ISL and English. The tool aims to bridge the language gap, providing ISL users (...) with an accessible and efficient solution that promotes inclusivity and autonomy. . (shrink)

Machine-based automation has long been a key factor in the modern era. However, lately, many people have been shocked by artificial intelligence (AI) applications, such as ChatGPT (OpenAI), that can perform tasks previously thought to be human-exclusive. With recent advances in natural language processing (NLP) technologies, AI can generate written content that is similar to human-made products, and this ability has a variety of applications. As the technology of large language models continues to progress by making (...) use of colossal reservoirs of digital information, AI is becoming more capable of recognizing patterns and associations within given contexts, making it especially helpful in assisting with various professional writing tasks [1]. In this paper, we will discuss several key points regarding the current state and future of AI in academia. Furthermore, since in certain aspects, acts can speak louder than words, we will also present some of our hands-on experiences of working with ChatGPT. (shrink)

The intersection of intelligent computing and linguistics has emerged as a vibrant field of study, offering innovative solutions and applications that transform how we understand and interact with language. This paper explores the diverse applications of intelligent computing in linguistics, encompassing natural language processing (NLP), computational linguistics, language modeling, speech recognition, and more. It delves into the underlying technologies, methodologies, and the impact of these advancements on various linguistic subfields. Through an extensive review of recent (...) literature, case studies, and practical examples, this paper aims to provide a comprehensive overview of the current state and future prospects of intelligent computing in linguistics. (shrink)

The current COVID-19 pandemic and the previous SARS/MERS outbreaks of 2003 and 2012 have resulted in a series of major global public health crises. We argue that in the interest of developing effective and safe vaccines and drugs and to better understand coronaviruses and associated disease mechenisms it is necessary to integrate the large and exponentially growing body of heterogeneous coronavirus data. Ontologies play an important role in standard-based knowledge and data representation, integration, sharing, and analysis. Accordingly, we initiated the (...) development of the community-based Coronavirus Infectious Disease Ontology in early 2020. -/- As an Open Biomedical Ontology (OBO) library ontology, CIDO is open source and interoperable with other existing OBO ontologies. CIDO is aligned with the Basic Formal Ontology and Viral Infectious Disease Ontology. CIDO has imported terms from over 30 OBO ontologies. For example, CIDO imports all SARS-CoV-2 protein terms from the Protein Ontology, COVID-19-related phenotype terms from the Human Phenotype Ontology, and over 100 COVID-19 terms for vaccines (both authorized and in clinical trial) from the Vaccine Ontology. CIDO systematically represents variants of SARS-CoV-2 viruses and over 300 amino acid substitutions therein, along with over 300 diagnostic kits and methods. CIDO also describes hundreds of host-coronavirus protein-protein interactions (PPIs) and the drugs that target proteins in these PPIs. CIDO has been used to model COVID-19 related phenomena in areas such as epidemiology. The scope of CIDO was evaluated by visual analysis supported by a summarization network method. CIDO has been used in various applications such as term standardization, inference, natural language processing (NLP) and clinical data integration. We have applied the amino acid variant knowledge present in CIDO to analyze differences between SARS-CoV-2 Delta and Omicron variants. CIDO's integrative host-coronavirus PPIs and drug-target knowledge has also been used to support drug repurposing for COVID-19 treatment. -/- CIDO represents entities and relations in the domain of coronavirus diseases with a special focus on COVID-19. It supports shared knowledge representation, data and metadata standardization and integration, and has been used in a range of applications. (shrink)

As if 2020 were not a peculiar enough year, its fifth month has seen the relatively quiet publication of a preprint describing the most powerful Natural Language Processing (NLP) system to date, GPT-3 (Generative Pre-trained Transformer-3), by Silicon Valley research firm OpenAI. Though the software implementation of GPT-3 is still in its initial Beta release phase, and its full capabilities are still unknown as of the time of this writing, it has been shown that this Artificial Intelligence (...) can comprehend prompts in natural language, on virtually any topic, and generate relevant, original text content that is indistinguishable from human writing. Moreover, access to these capabilities, in a limited yet worrisome enough extent, is available to the general public as of the time of this writing. This paper presents select examples of original content generated by the author using GPT-3. These examples illustrate some of the capabilities of GPT-3 in comprehending prompts in natural language and generating convincing content in response. We use these examples to raise specific, fundamental questions pertaining to the intellectual property of this content and the potential use of GPT-3 to facilitate plagiarism. Our goal is to instigate not just a sense of urgency, but of a present tardiness on the part of the academic community in addressing these questions. (shrink)

This paper investigates how large language models (LLMs) can be applied to publicly available financial data to generate automated financial summaries and provide actionable recommendations for investors. We demonstrate how LLMs can process both structured financial data (balance sheets, income statements, stock prices) and unstructured text (earnings calls, management commentary) to derive insights, predict trends, and automate financial reporting. By focusing on a specific publicly traded company, this research outlines the methodology for leveraging LLMs to analyze company performance and (...) generate investor-focused summaries and recommendations. (shrink)

Consultations are key to gather evidence that informs rulemaking. When analysing the feedback received, it is essential for the regulator to appropriately cluster stakeholders’ opinions, as misclustering may alter the representativeness of the positions, making some of them appear majoritarian when they might not be. The European Commission (EC)’s approach to clustering opinions in consultations lacks a standardized methodology, leading to reduced procedural transparency, while making use of computational tools only sporadically. This paper explores how natural language (...) class='Hi'>processing (NLP) technologies may enhance the way opinion clustering is currently conducted by the EC. We examine 830 responses to three legislative proposals (the Artificial Intelligence Act, the Digital Markets Act and the Digital Services Act) using both a lexical and semantic approach. We find that some groups (like small and medium companies) have low similarity across all datasets and methodologies despite being clustered in one opinion group by the EC. The same happens for citizens and consumer associations for the consultation run over the DSA. These results suggest that computational tools actually help reduce misclustering of stakeholders’ opinions and consequently allow greater representativeness of the different positions expressed in consultations. They further suggest that the EC could identify a convergent methodology for all its consultations, where such tools are employed in a consistent and replicable rather than occasionally. Ideally, it should also explain when one methodology is preferred to another. This effort should find its way into the Better Regulation toolbox (EC 2023). Our analysis also paves the way for further research to reach a transparent and consistent methodology for group clustering. (shrink)

The present comment concerns a recent AI & Ethics article which purports to report evidence of speciesist bias in various popular computer vision (CV) and natural language processing (NLP) machine learning models described in the literature. I examine the authors' analysis and show it, ironically, to be prejudicial, often being founded on poorly conceived assumptions and suffering from fallacious and insufficiently rigorous reasoning, its superficial appeal in large part relying on the sequacity of the article's target readership.

As the demand for explainable deep learning grows in the evaluation of language technologies, the value of a principled grounding for those explanations grows as well. Here we study the state-of-the-art in explanation for neural models for natural-language processing (NLP) tasks from the viewpoint of philosophy of science. We focus on recent evaluation work that finds brittleness in explanations obtained through attention mechanisms.We harness philosophical accounts of explanation to suggest broader conclusions from these studies. From this (...) analysis, we assert the impossibility of causal explanations from attention layers over text data. We then introduce NLP researchers to contemporary philosophy of science theories that allow robust yet non-causal reasoning in explanation, giving computer scientists a vocabulary for future research. (shrink)

Efficient rake scheduling and demand forecasting in railway operations are essential to address the complexities of passenger demand, minimize delays, and enhance utilization. This project uses advanced machine learning methods, specifically LSTM (Long Short-Term Memory) networks and GBM (Gradient Boosting Machine), to predict demand and optimize rake scheduling dynamically. Integrating a user-friendly web interface allows realtime data monitoring, enabling railway operators to make informed decisions. By leveraging real-time data sources, including rake movement, schedules, weather, and traffic conditions, this project aims (...) to improve operational efficiency and responsiveness. Keywords. Railway Scheduling, Rake Optimization, Demand Forecasting, LSTM, GBM, Machine Learning, RealTime Data. (shrink)

News Rakshak is an advanced web application that monitors, analyzes, and verifies news related to government schemes in regional media across multiple platforms, including 200+ websites, e-papers, and YouTube channels. The application leverages Natural Language Processing (NLP) for sentiment analysis and integrates Artificial Intelligence (AI) to identify and flag fake news and videos. Sentiments are classified as positive, neutral, or negative, rated on a scale from 0 to 5, providing Public Information Bureau (PIB) officers with actionable insights. (...) The system uses Optical Character Recognition (OCR) for extracting text from image-based news formats and delivers real-time alerts to officials via multiple channels. The user-friendly interface, equipped with filtering options, enables effective trend visualization. News Rakshak serves as a comprehensive tool for tracking and verifying government-related news sentiment and authenticity, ensuring accurate feedback from regional media. (shrink)

Students often struggle to capture accurate and comprehensive notes during lectures, which is crucial for effective learning. Traditional note-taking methods are time-consuming and prone to errors, especially when lectures include digressions or complex topics. Existing solutions, like manual transcription or basic audio recordings, fall short as they requireextensive post-processing and fail to structure information efficiently. This project proposes an automated Audio-to-Notes Summarization System that converts classroom audio into concise, structured notes. By leveraging advanced speech recognition, Natural Language (...) processing (NLP), and machine learning, the system filters out non essential content, focusing on core subject matter. This approach aims to enhance student learning by providing accurate, easily accessible notes, thus improving academic performance and reducing the time spent on note-taking. (shrink)

The "Udyog Saarthi App" is a progressive web-based application developed to enhance job coaching and career readiness for adults with disabilities, specifically focusing on the 4% reservation opportunities provided by NIEPMD and similar institutions in India. This innovative platform addresses the unique needs of users by allowing them to select their specific disability during registration, which customizes the website’s layout for improved accessibility. The application offers a comprehensive listing of real-time job opportunities that align with users' skills and qualifications, empowering (...) them to pursue suitable career paths. In addition, it provides personalized learning resources tailored to each job role, ensuring that users are equipped with the necessary skills to excel. The app also includes periodic mock assessments designed to evaluate users' preparedness for employment. These assessments employ natural language processing (NLP) techniques to calculate scores based on various parameters such as readability scores and the SMOG index, promoting objective and effective evaluations. Moreover, the Udyog Saarthi App features an interactive dashboard that enables users to track their progress and analytics, fostering a sense of accomplishment and motivation. To promote community engagement and support, the application includes a friend recommender system that connects users with similar disabilities, facilitating peer networking and encouragement. Incorporating multilingual support through the Google Translate API, the platform aims to reach a diverse user base, ensuring that language barriers do not hinder accessibility. Ultimately, the Udyog Saarthi App strives to empower individuals with disabilities by providing essential tools and resources, enabling them to become self-sufficient and industry-ready in the competitive job market. (shrink)

At the heart of natural language processing is the understanding of context dependent meanings. This paper presents a preliminary model of formal contexts based on situation theory. It also gives a worked-out example to show the use of contexts in lifting, i.e., how propositions holding in a particular context transform when they are moved to another context. This is useful in NLP applications where preserving meaning is a desideratum.

Sự tiến bộ nhanh chóng của các nền tảng Công nghệ Thông tin (CNTT) và ngôn ngữ lập trình đã làm thay đổi hình thái vận động và phát triển của xã hội loài người. Không gian mạng và các tiện ích đi kèm ngày càng được mở rộng, dẫn đến sự chuyển dịch dần từ đời sống trong thế giới thực sang đời sống trong thế giới ảo (còn gọi là không gian mạng hay không gian số). Sự mở (...) rộng và phát triển của các Mô hình Xử lý Ngôn ngữ Tự nhiên (Natural Language Processing – NLP) và Mô hình Ngôn ngữ Lớn (Large Language Model – LLM) thể hiện những đặc điểm giống con người về lập luận, nhận thức, sự chú ý và sáng tạo đã giúp con người vượt qua rào cản về khả năng vận hành. Đi kèm với các tiềm năng to lớn của AI là những lỗ hổng bảo mật mới và các rủi ro về an ninh thông tin phức tạp hơn. Khi xã hội vẫn đang cố gắng chuyển mình sang giai đoạn mới để có thể thích ứng với sự biến đổi công nghệ, cuộc cách mạng AI lại tiếp tục diễn ra đòi hỏi chúng ta phải suy nghĩ lại quỹ đạo của quá trình chuyển đổi xã hội vì nó có thể làm trầm trọng hơn các rủi ro an ninh thông tin được trình bày ở trên. Cụ thể, xã hội cần tiến hóa như thế nào để có thể bắt kịp với sự biến đổi mang tính đột phá tạo ra bởi làn sóng công nghệ AI hiện nay? Làm sao để quản lý và tận dụng sức mạnh của chúng trong khi vẫn đảm bảo được an ninh thông tin khi không gian và thời gian sống trong thế giới ảo của chúng ta ngày càng tăng? Bài viết này sẽ giúp làm sáng tỏ và trả lời các câu hỏi này. (shrink)

P&C insurers have an important role in addressing financial risk management needs but now struggle to respond to the new forms of risk. Historical analysis and actuarial calculations, which form the backbone of classical approaches to risk measurement and management, are not well suited to such new kinds of risks as climate change, cyber risks, and business cycle risks. These conventional approaches are also a static method for selling, which has limited potential in changing quickly with new market and consumer (...) changes. However, the development of AI, big data, or advanced predictive analytics has provided a new paradigm to such challenges to cope with and analyze big data dynamically to derive Applying a combination of best Machine Learning (ML), natural language processing (NLP) and other AI techniques, it will become possible to predict risks exposures, as well as optimize underwriting and claims processing. In this paper, the author investigates the opportunities that AI carries within the sphere of reformation of risk evaluation approaches in P&C insurance. In particular, it discusses the ever-emerging field with AI models integrated into operations, underwriting, fraud detection, claims modelling, and loss computation. The applied research uses a comprehensive bibliographical review to present past and present practices and assess modern AI implementations. This paper uses research approaches and case studies to prove that AI can solve some problems and bring cost-saving solutions. Moreover, it emphasizes the ability of predictive analytics to enhance customer satisfaction at the policy level as well as optimize operations. However, the paper also explores the dirty side of implementation, including the stipulations surrounding compliance and data privacy and thus provides an objective analysis of the future of AI use in the P&C Insurance Industry. (shrink)

The Grievance Management System is a web-based platform designed to enhance the efficiency and transparency of handling public grievances. By introducing role-based access for users, moderators, and government officials, the system ensures that grievances are systematically reviewed, prioritized, and resolved. Users can submit grievances, track their status, and receive notifications regarding updates. Moderators are tasked with verifying the validity of each grievance and assigning it a priority level before passing it on to government officials for action. Government officials, in turn, (...) have the ability to filter grievances by priority and submission date, helping them manage and resolve issues more effectively. The system integrates technologies such as Flask for backend logic, MySQL for database management, and natural language processing (NLP) to classify grievances into relevant departments. It also prevents rejected grievances from being re- submitted and delivers real-time notifications to users, ensuring an efficient and transparent grievance handling process. This project aims to improve the accountability and responsiveness of government departments in addressing public complaints. (shrink)

In the era of big data and distributed ecosystems, understanding the origin, flow, and transformation of data across complex infrastructures is critical for ensuring transparency, accountability, and informed decision-making. As data-driven enterprises increasingly rely on hybrid cloud architectures, data lakes, and real-time pipelines, the complexity of tracking data movement and transformations grows exponentially. Traditional data lineage solutions, often based on static metadata extraction or rule-based approaches, are insufficient in dynamically evolving environments and fail to provide granular, context-aware insights. This research (...) introduces an AI-augmented, cognitive graph-based framework for autonomous data lineage, designed to enhance data traceability in large-scale and heterogeneous data ecosystems. The framework leverages cutting-edge machine learning (ML), natural language processing (NLP), and graph-based reasoning techniques to enable intelligent discovery, semantic interpretation, and continuous monitoring of data assets throughout their lifecycle. The core of the solution lies in the construction of a dynamic cognitive graph that represents relationships between datasets, processes, systems, and users, enriched with contextual annotations and temporal dimensions. Our architecture incorporates self-learning mechanisms, enabling adaptive lineage discovery and automated anomaly detection. By applying reinforcement learning and stream analytics, the framework not only maps data flows in real time but also evolves with system changes, schema variations, and business logic updates. It provides both forward and backward traceability, supports impact analysis, and enhances compliance auditing capabilities. Furthermore, our system is capable of processing both structured and unstructured metadata, employing advanced NLP models to extract implicit lineage information from data dictionaries, SQL queries, and documentation. The result is a holistic, intelligent, and scalable data lineage solution that reduces manual intervention, mitigates operational risks, and supports regulatory compliance frameworks such as GDPR, HIPAA, and SOX. Experimental evaluations conducted in hybrid cloud environments using tools such as Apache Kafka, Neo4j, Spark, and Python-based ML libraries demonstrate a significant improvement in lineage coverage, anomaly detection accuracy, and system scalability. Compared to conventional lineage tools, our AI-augmented framework delivers a 30% increase in traceability precision and a 40% reduction in manual effort required for lineage tracking and governance. This research lays the foundation for a new paradigm in data governance, where AI not only enhances observability but enables autonomous cognition within data infrastructure. The framework is poised to play a critical role in enabling data democratization, operational agility, and enterprise-wide data literacy. (shrink)

The early detection of cognitive concerns is crucial for timely intervention and improved patient outcomes. However, analyzing large-scale real-world data for cognitive decline presents significant challenges in efficiency and accuracy. This paper introduces an Autonomous AI Framework that leverages machine learning and natural language processing (NLP) to identify cognitive concerns from diverse datasets, including electronic health records (EHRs), social media interactions, and clinical notes. Our approach integrates deep learning models, feature selection techniques, and interpretability methods to enhance (...) detection accuracy and provide actionable insights. Experimental results demonstrate the framework’s effectiveness in identifying early signs of cognitive issues with high precision and recall. (shrink)

Sự tiến bộ nhanh chóng của các nền tảng Công nghệ Thông tin (CNTT) và ngôn ngữ lập trình đã làm thay đổi hình thái vận động và phát triển của xã hội loài người. Không gian mạng và các tiện ích đi kèm ngày càng được mở rộng, dẫn đến sự chuyển dịch dần từ đời sống trong thế giới thực sang đời sống trong thế giới ảo (còn gọi là không gian mạng hay không gian số). Sự mở (...) rộng và phát triển của các Mô hình Xử lý Ngôn ngữ Tự nhiên (Natural Language Processing – NLP) và Mô hình Ngôn ngữ Lớn (Large Language Model – LLM) thể hiện những đặc điểm giống con người về lập luận, nhận thức, sự chú ý và sáng tạo đã giúp con người vượt qua rào cản về khả năng vận hành. Đi kèm với các tiềm năng to lớn của AI là những lỗ hổng bảo mật mới và các rủi ro về an ninh thông tin phức tạp hơn. Khi xã hội vẫn đang cố gắng chuyển mình sang giai đoạn mới để có thể thích ứng với sự biến đổi công nghệ, cuộc cách mạng AI lại tiếp tục diễn ra đòi hỏi chúng ta phải suy nghĩ lại quỹ đạo của quá trình chuyển đổi xã hội vì nó có thể làm trầm trọng hơn các rủi ro an ninh thông tin được trình bày ở trên. Cụ thể, xã hội cần tiến hóa như thế nào để có thể bắt kịp với sự biến đổi mang tính đột phá tạo ra bởi làn sóng công nghệ AI hiện nay? Làm sao để quản lý và tận dụng sức mạnh của chúng trong khi vẫn đảm bảo được an ninh thông tin khi không gian và thời gian sống trong thế giới ảo của chúng ta ngày càng tăng? Phần một của bài viết sẽ bàn về các vấn đề và rủi ro ảnh hưởng đến an ninh thông tin trong kỷ nguyên AI, cũng như là vai trò, lợi thế, và cơ hội của việc vận dụng AI cho mục đích đảm bảo an ninh thông tin. (shrink)

The increasing demand for secure and efficient voting systems has led to the exploration of online voting solutions. Traditional voting methods are often vulnerable to fraud, inefficiencies, and logistical challenges. This paper presents an online voting system that leverages machine learning techniques to enhance security, accuracy, and accessibility. The system employs facial recognition for voter authentication, anomaly detection to prevent fraudulent activities, and natural language processing (NLP) for user interaction. Experimental results indicate that the proposed model provides (...) a reliable, tamper-resistant, and user-friendly voting experience. (shrink)

The rapid advancement of artificial intelligence (AI) and machine learning (ML) has led to significant innovations in the healthcare sector. One such development is AI-powered healthcare chatbots, which assist patients and medical professionals by providing medical guidance, symptom assessment, and appointment scheduling. This paper presents the design and implementation of an AI healthcare chatbot using machine learning techniques. The chatbot leverages natural language processing (NLP) and deep learning models to understand and respond to user queries effectively. Experimental (...) results demonstrate the chatbot's efficiency in delivering accurate and context-aware medical information. (shrink)

Ever since the era of internet had ushered in cloud computing, there had been increase in the demand for the unlimited data available through cloud computing for data analysis, pattern recognition and technology advancement. With this also bring the problem of scalability, efficiency and security threat. This research paper focuses on how data can be dynamically mine in real time for pattern detection in a secure cloud computing environment using combination of decision tree algorithm and Random Forest over a restful (...) Application Programming Interface (API). We are able to successfully Implement data mining on cloud computing bypassing or avoiding direct interaction with data warehouse and without any terminal involve by using combination of IBM Cloud storage facility, Amazing Web Service, Application Programming Interface and Window service along with a decision tree and Random Forest algorithm for our classifier. We were able to successfully bypass direct connection with the data warehouse and cloud terminal with 94% accuracy in our model. (shrink)

Estimating word complexity is a well-established task in computer-assisted language learning. So far, however, complexity estimation has been largely limited to comprehension. This neglects words that are easy to comprehend, but hard to produce. We introduce semantic error prediction (SEP) as a novel task that assesses the production complexity of content words. Given the corrected version of a learner-produced text, a system has to predict which content words replace tokens from the original text. We present and analyse one example (...) of such a semantic error prediction dataset, which we generate from an error correction dataset. As neural baselines, we use BERT, RoBERTa, and LLAMA2 embeddings for SEP. We show that our models can already improve downstream applications, such as predicting essay vocabulary scores. (shrink)