Browse Items (16488 total)
Sort by:
-
An Analysis of Word Sense Disambiguation (WSD)
Word sense disambiguation (WSD) is the method of using computer algorithms to determine the sense of arguments in the background. As a result of its difficult nature, WSD has measured an AI-complete problem, i.e., a problem whose key is as minimum as difficult as those posed by artificial intelligence. This article describes the task and introduces motives to resolve the ambiguity of words discussed throughout the text. This article summarizes supervised, unsupervised, and knowledge-based solutions. Senseval/semeval campaigns are described in relation to the assessment of WSDs, with the aim of an unbiased assessment of schemes working on numerous disambiguation errands. Finally, future directions, requests, open difficulties, and open problems are discoursed. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
A Novel Steganographic Approach for Image Encryption Using Watermarking
Steganography is a technique for obfuscating secret information by enclosing it in a regular, non-secret file or communication; the information is subsequently extracted at the intended location. Steganography can be used in addition to encryption to further conceal or safeguard data. Watermarking is one such technique practiced in the area of steganography. Watermarking can be practiced via multiple algorithmic techniques like Discrete Wavelength Transform (DWT), Discrete Cosine Transform (DCT), Singular Value Decomposition (SVD), Discrete Fourier Transform (DFT). In this study, a combination of such approaches along with AES encrypted watermarked images has been implemented. Validation of these techniques has been achieved by evaluating the Peak Signal to Noise Ratio (PSNR). 2023 IEEE. -
Exploring the Influence of Service Learning on the Socio-Educational Commitment and Self- Efficacy of Graduate Educators in the Artificial Intelligence (AI) Domain.
This study, conducted by a distinguished university, aims to contribute significantly to the professional development of educators dedicated to creating a fair, sustainable, and socially conscious world. The research focuses on a pedagogical approach using Service Learning to foster civic and social skills in higher education students. The main goal is to examine how graduate students, actively participating in Service-Learning initiatives, develop socio-educational commitment and self-efficacy compared to traditional university volunteering. The study, involving 1562 aspiring educators, employs a quantitative correlational methodology. The hypothesis suggests that Service-Learning leads to more positive outcomes in socio-educational commitment, pedagogical self-efficacy, and crafting instructional materials. The findings, statistically significant (p < 0.01), highlight the increased development of these metrics among participants in Service-Learning programs. 2024 IEEE. -
AI as sustainable and eco-friendly environment for climate change
[No abstract available] -
Feature extraction and fusion techniques for multimodal data
Integrating multimodal information has become crucial in the big data era for developing a thorough knowledge of complex systems and enhancing decision-making in a variety of fields. The importance of feature extraction and fusion strategies in multimodal learning is examined in this chapter, with particular attention paid to the difficulties and approaches involved in merging several data modalities, including text, pictures, audio, and sensor data. It talks about how conventional feature extraction approaches have evolved into more sophisticated ones like deep learning models for picture and audio data and neural embeddings for text. The chapter also explores several fusion tactics, such as early, late, and intermediate fusion, and focuses on how they are used in domains including sentiment analysis, autonomous cars, healthcare, and multimodal search engines. The chapter highlights future directions, such as lightweight architectures and privacy-preserving techniques, while also addressing contemporary issues, such as managing missing data, scalability, and privacy concerns. The chapter provides a thorough grasp of how feature extraction and fusion aid in the creation of multimodal systems that are more precise, effective, and interpretable by looking at these factors. 2026 Elsevier Inc. All rights reserved. -
Anti-Semitic Content on Social Media Analyses Using a Hybrid Model
In the modern world, people have a variety of channels to freely express their opinions, thoughts, knowledge, and feelings on many topics on social media. However, they abuse this freedom by spewing hate speech based on a person's or group's ethnicity, gender, or religion, caste, sexual orientation, ethnicity, and nationality. Hate speech is the most likely form of expression for hostility and superstition on social media. One of the causes of cyberbullying, which can have an effect on social life on both a national and personal level, is an increase in hate speech. Hateful material has the ability to hurt and an individual and promote social unrest. Social media platforms are unable to monitor every topic that is posted by users, so automated detection of hate speech is a crucial tool. Utilizing machine learning models was first popular. But as the dataset size grows, these models are unable to deliver adequate results. While advanced deep learning models such as Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional Encoder Representations from Transformers (BERT), and similar architectures have demonstrated greater reliability and effectiveness. Toxic speech is a typical thing which is floating around on social media in this fast-paced world where social media is a significant part of our lives and influences the thoughts of many people, where people have access to distribute whatever sort of information they want. Fighting this is difficult because hate speech recognition is now essential in modern society. In order to extract the necessary features and determine if a section of news contains hate speech or not, this article uses a hybrid CNN+LSTM architecture. 2025 IEEE. -
Case studies: multimodal applications in natural language processing
This chapter explores the incorporation of natural language processing (NLP) with multimodal information sources, including text, speech, and visual information, towards the improvement of practical applications. NLP may very significantly enhance tasks such as sentiment analysis, image captioning, and cross-modal retrieval by combining these modalities. Two examples of deep learning approaches are neural networks and transformers, which are examples of critical approaches for developing robots that analyze and understand complex multimodal inputs. The chapter is full of case examples that illustrate how multimodal NLP can revolutionize many industries, including healthcare data analysis and voice-activated assistant development. These illustrations demonstrate how NLP can enhance user interactions and decision-making processes by offering deeper, more contextual insights. In fact, the chapter also covers issues and ways ahead of multimodal NLPas integrating data, handling faulty or missing data, and how to resolve ethical dilemmas. These ongoing changes will define future artificial intelligence systems with increased adaptability, intuitiveness, and applicability. 2026 Elsevier Inc. All rights reserved. -
Study of Optimization Techniques in Agriculture
In agriculture, optimization strategies are essential for raising production, sustainability, and resource efficiency. This abstract explores several agricultural optimization approaches and highlights their importance in contemporary farming operations. Obstacles to traditional agricultural operations include climate change, resource constraint, and shifting consumer preferences. Through the application of cutting-edge technologies and rigorous scientific methods, optimization strategies provide answers to these problems. Precision agriculture, which uses data-driven techniques like remote sensing, Geographic Information System (GIS), and Global Positioning System (GPS) to customize farming procedures to particular field conditions, is one important area of optimization. This helps farmers to maximize yields while minimizing waste and the environmental effect of inputs like water, fertilizer, and pesticides. In addition, optimization methods include selective breeding for genetic advancements and biotechnology, with the goal of creating crops with increased nutritional value, disease resistance, and production potential. Under the category of optimization approaches are integrated pest management solutions, which efficiently control pests and illnesses while using less chemical pesticides by applying ecological principles. In summary, agricultural optimization approaches offer a comprehensive strategy to tackle the issues that contemporary farming faces, encouraging resilience, productivity, and sustainability in food production systems. Given the changing global agricultural dynamics, it is imperative that these strategies be further researched and used in order to ensure environmental stewardship and food security. 2026 Scrivener Publishing LLC. All rights reserved. -
Navigating the ethical landscape of artificial intelligence: Challenges, frameworks, and responsible deployment
In artificial intelligence (AI), machine learning (ML) has become a game-changing concept that allows systems to learn from experience and get better without explicit programming. This chapter explores the main ideas, techniques, and applications of ML, offering a succinct introduction to the field. The first step in the process is to gain a basic understanding of supervised learning, which is the process by which algorithms learn to make predictions or judgements from labelled training data. Next, we introduce unsupervised learning, which emphasizes finding patterns in unlabelled data and frequently results in interesting findings and clustering. To emphasize the importance of reinforcement learning in decision-making processes, the paradigm is presented where agents learn by interacting with an environment and receiving feedback. Ideas related to ML, such as feature engineering, model assessment, and the balance between variance and bias, are discussed. The significance of quality data in ML applications is emphasized, along with the impact of data pretreatment on model performance. It also clarifies how neural networks, a branch of ML, simulate the workings of the human brain. The ability of deep learning, a branch of ML that makes use of multi-layered neural networks, to handle challenging tasks such as speech and picture recognition is being investigated. In order to emphasize the necessity of responsible ML model deployment and usage, practical factors are emphasized, including the significance of ethical considerations and responsible AI. The final section of the chapter offers a preview of MLs future, discussing issues and trends that practitioners and researchers should be aware of. This chapter essentially functions as a thorough introduction to ML principles, providing an overview of the wide range of ML approaches, applications, and ethical issues that support the technologys transformative potential across a range of industries. 2025 selection and editorial matter, G. Sucharitha, Anjanna Matta, M. Srinivas and Sachi Nandan Mohanty; individual chapters, the contributors. -
Explainable AI in Healthcare: A Hybrid CNN-ViT Approach for Pneumonia Detection Using SHAP
The adoption of Artificial Intelligence (AI) in healthcare has improved diagnostic accuracy, particularly in medical imaging. However, the opaque nature of deep learning models raises concerns about interpretability in high-stakes applications such as pneumonia diagnosis. This study proposes an Explainable AI (XAI) framework that integrates a Hybrid CNN-ViT architecture with SHAP (SHapley Additive Explanations) for pneumonia detection from chest X-rays. Our approach achieves competitive diagnostic performance (94% accuracy) while enhancing transparency by highlighting clinically relevant features such as lobar consolidations and ground-glass opacities. By grounding explanations in established radiological findings, the framework supports clinical trust and regulatory compliance. This work contributes to bridging the gap between AI performance and medical accountability, positioning explainable deep learning as a trustworthy tool for real-world healthcare deployment. The Author(s), under exclusive license to Springer Nature Switzerland AG 2026. -
Challenges in preprocessing and normalization of heterogenous data
In todays information-driven landscape, the exposure to information from a wide range of sources, such as social media, financial transactions, healthcare records, and Internet of Things (IoT) sensors. This variety, while providing valuable insights, also brings significant issues. Heterogeneous information, which varies in formats, structures, and scales, requires careful management to ensure it can be effectively used in analytics and machine learning. Key steps in this process include prehandling and standardization. Prehandling involves cleaning and preparing the information by tackling issues like missing values, identifying and eliminating noise (inaccurate or irrelevant information), and addressing inconsistencies. Normalization, in contrast, converts the information into a uniform format and scale, facilitating easier comparison and analysis. However, managing diverse information effectively comes with several issues. Missing information is a frequent problem, and accurately filling in these gaps can be complicated, especially for complex information types. Noise and inconsistencies can greatly affect the accuracy and reliability of any analysis that follows. Additionally, merging information from various sources with differing formats and structures can be a challenging and time-consuming task. This chapter explores the specific issues faced when prehandling and normalizing different information types, including numerical, categorical, textual, and image information. Real-world examples from India, such as the Aadhaar information base and IoT-enabled smart cities, highlight the practical implications of these issues. By grasping best practices and emerging AI-driven trends, organizations can improve information reliability and enhance decision-making. 2026 Elsevier Inc. All rights reserved. -
Social Inclusion and Sustainability
Sustainability and social inclusion are closely intertwined concepts, both centered on ensuring equal access to opportunities, resources, and participation in decision-makingwhile also committing to the long-term protection of the environment. The essay describes how social inclusion catalyzes sustainable development; more specifically, it illuminates why the inclusion of marginalized groups in economic and environmental policy is so crucial. Including marginalized groups in sustainable development processes catalyzes a variety of views and promotes social justice, thereby maximizing the effectiveness of efforts toward sustainability. Access to resources, employment opportunities, and education go hand in hand in building a just and fair society. Sustainable job opportunities are tied to environmental goals and narrow the economic gap. Equal access to education also allows everybody, regardless of their background, to join in and benefit from the environmentally sound practices. To minimize the inequalities associated with environmental degradation and climatic change impacts on vulnerable populations, environmental justice is key to social inclusion, as every citizen should be represented through buildings that cause minimum damage to the environment. There should be an investment in inclusive urban development, sustainable housing, and infrastructure for human settlement. This paper emphasizes that social inclusion and sustainability goals harmoniously balance each other in the pursuit of an equitable, resilient, and ecologically responsible future. 2026 John Wiley & Sons Ltd. All rights reserved. -
Transfer learning in multimodal settings
A powerful machine learning technique in the multimodal environment allows the transmission learning model to adapt to information from one domain to another, which promotes more effective learning in different types of data, including lessons, images, speeches, speeches, and sensor data. This method increases the model's adaptability, reduces the requirement for large marked datasets, and increases performance across domains. It has been used in several domains where multimodal integration is important, such as healthcare, autonomous systems, and natural language processing. Despite the benefits, transmission learning has disadvantages, including high data costs, data shortages, and domain changes. To meet these challenges, model architecture, adaptation strategy, and improvement in dataset growth techniques are necessary. This study examines basic ideas, procedures, and transfer of transfer to multimodal references, and provides insight. Practical use and new development. We show the developing role to learn transfer in improving artificial intelligence (AI) applications by looking at current studies and case studies. As the area develops, a combination of knowledge from many methods will be necessary to create scalable, reliable, and effective AI systems that can handle the problems in the real world. 2026 Elsevier Inc. All rights reserved. -
Multimodal data analytics for social media and user behavior
The introduction of social media has prompted an explosion of diverse data types, such as textual content, pix, videos, and audio. Traditional unimodal analysis techniques do not effectively depict the difficult interactions between exceptional fact sorts and consumer sports. Multimodal data analytics addresses this difficulty through fusing one-of-a-kind modalities to unlock deeper insights, improving the accuracy and scope of social media analysis. This bankruptcy looks into the significance of multimodal facts in understanding consumer behavior, sentiment analysis, content material engagement assessment, and trend prediction. The bankruptcy starts off with the exploration of various sources of statistics in social media analytics, together with textual content posts, visual content, and consumer interactions. It then explores preprocessing and function extraction strategies utilized to prepare raw multimodal data for the usage of gadget gaining knowledge of. In-intensity methodologies, inclusive of natural language processing for text evaluation, computer vision for photo and video interpretation, and speech recognition for audio processing, are expounded in extraordinary detail. Integration of these modalities via fusion techniquesearly fusion, past due fusion, and hybrid modelsis also explored. 2026 Elsevier Inc. All rights reserved. -
Challenges and opportunities in multimodal learning research
The trend of multimodal learning, which involves processing and interpreting data through multiple modes such as text, images, and audio, is one aspect that highlights a great frontier in the artificial intelligence (AI) and machine learning (ML) domains. This project explores the technical, practical, and ethical considerations of research studies on multimodality. It starts with preliminary ethical considerations that should drive progress in AI and ML technologies, which include ideas of transparency, accountability, equality, and privacy. Analysis of this paper holds prime importance for moral concerns, as it discusses the issues of bias in AI algorithms and gives strategies that may reduce the level of bias in multimodal patterns. This technology part of the research focuses on technical challenges concerning accountability and transparency in multimodal machine decision-making methodologies. Privacy concerns regarding extensive use of AI and ML have been brought forward, along with the strategy for defensive personal statistics. At each step, an opportunity for innovation and development will be sought and mapped through the complex ethical landscapes of multimodal knowledge research. Through these considerations, this observation attempts to provide a close analysis in which future recommendations and discussions in the realm of AI and multimodal learning are addressed. 2026 Elsevier Inc. All rights reserved. -
Family Factors Associated with Problematic Use of the Internet in Children: A Scoping Review
Background: Problematic use of the internet (PUI) is a growing concern, particularly in the young population. Family factors influence internet use among children in negative ways. This study examined the existing literature on familial or parental factors related to PUI in children. Methods: A scoping review was conducted in EBSCOhost, PubMed, ScienceDirect, JSTOR, Biomed Central, VHL Regional Portal, Cochrane Library, Emerald Insight, and Oxford Academic Journal databases. Studies reporting data on family factors associated with PUI in children, published in English in the 10 years to July 2020 were included. The following data were extracted from each paper by two independent reviewers: methodology and demographic, familial, psychiatric, and behavioral correlates of PUI in children. Results: Sixty-nine studies fulfilled the eligibility criteria. Three themes emerged: parenting, parental mental health, and intrafamilial demographic correlates of PUI in children. Parenting styles, parental mediation, and parentchild attachment were the major parenting correlates. Conclusion: Literature on significant familial and parental factors associated with PUI in children is scarce. More research is required to identify the interactions of familial and parental factors with PUI in children, to develop informed management strategies to address this issue. 2022 Indian Psychiatric Society - South Zonal Branch. -
Application of CNN and Recurrent Neural Network Method for Osteosarcoma Bone Cancer Detection
The outlook for people with metastatic osteosarcoma at an advanced stage is poor. Osteosarcoma is the most frequent form of bone cancer in children and young adults. There is an urgent need for both advances in treatment tactics and the identification of novel therapeutic targets for osteosarcoma since the disease typically develops resistance to existing treatments. Cancer stem cells, also known as tumor stem cells, have been linked to the development and spread of cancer at multiple points in the disease's progression. Cancer stem cells are linked to treatment resistance and carcinogenesis, and recent studies have demonstrated that osteosarcoma shares these properties. The proposed methodology rests on the three pillars of preprocessing, feature extraction, and model training. During preprocessing, that the proposed approach eliminated isolated highlights to help us zero in on the trustworthy region. They use the wavelet transform and the gray level co-occurrence matrix to extract features. A CNN-RNN technique is used to evaluate the models. In terms of output quality, the proposed technique is superior to both CNN and RNN. 2023 IEEE. -
Improved image denoising with the integrated model of Gaussian filter and neighshrink SURE
Image denoising, being an important preprocessing stage in image processing, minimizes the noise interfering with the information content of the image. The denoising problems are addressed by various techniques starting from the Fourier transforms to wavelets. Because of the localized time frequency features and advantages of multi resolution capabilities, the wavelets have been extensively used in the denoising process. The development of algorithms for the wavelet thresholding or shrinkage strategies along with different filters have resulted in the betterment of image quality after the denoising. Even though the image denoising algorithm based on a combination of Gaussian and Bilateral filters, shows good performance but lacks in consistency with respect to the noise levels and also the type of images used. This paper discusses the advantages of NeighShrink SURE rule in developing an effective thresholding strategy, thereby proposing a improved denoising method incorporating the NeighShrink SURE rule along with combination of Gaussian filter model. The methodology employs the use of subband thresholding derived from the NeighShrink SURE rule. The outcome of the proposed method exhibits a comparatively improved performance in Peak Signal to Ratio (PSNR) and Image Quality Index (IQI) values of the test images. BEIESP. -
An efficient image denoising method based on bilateral filter model and neighshrink SURE
In all the instances of image acquisition, transmission and storage, the unwanted noise gets into the information content of the image and thereby introduces an unpleasant visual quality to the observer. So the field of image processing has produced a lot of image denoising algorithms and techniques to improve the visual quality of the image. Since noise cannot be reduced to zero practically, the need for faithful and efficient denoising techniques to produce almost noiseless images demands a systematic research work in the field of denoising methods. The denoising process using a bilateral filter even though produces improvement in the image quality, it does not show consistency when the noise level is high and also the peak signal to noise ratio (PSNR) and Image quality Index (IQI) do not show any improvement. This paper proposes an improved algorithm that incorporates the function of bilateral filter model and wavelet thresholding using Neighshrink SURE method. The results show significant improvement in both PSNR and IQI values with respect to the four standard test images under various noise conditions. BEIESP.