Browse Items (16488 total)
Sort by:
-
ML based sign language recognition system
This paper reviews different steps in an automated sign language recognition (SLR) system. Developing a system that can read and interpret a sign must be trained using a large dataset and the best algorithm. As a basic SLR system, an isolated recognition model is developed. The model is based on vision-based isolated hand gesture detection and recognition. Assessment of ML-based SLR model was conducted with the help of 4 candidates under a controlled environment. The model made use of a convex hull for feature extraction and KNN for classification. The model yielded 65% accuracy. 2021 IEEE. -
ML in drug delivery-current scenario and future trends
Machine learning (ML) has enabled transformative applications and emerged as a domain-agnostic decision-making tool as a virtue of its rapid democratization. The authors believe that a systematic assortment of important publications on this issue is indispensable in this context. In terms of data ingestion, data curation, data preprocessing, data handling, and model cross-validation, this review gathers together several studies that have demonstrated a minimum ML framework approach. In general, ML models are described as black-box models, with limited information supplied about their transparency. The authors propose techniques based on the US Food and Drug Administration (FDA)'s current good ML practice (GMLP) in order to improve the ML framework and minimize the aforementioned gap, especially for data. Considering this, the conversation around a model's logic and interpretability are additionally provided. Explicitly, the authors explore the challenges and constraints that ML execution confronts throughout the development of pharmaceuticals. In this context, a structural approach in statistics is presented to allow the scientist to assess the quality of data and incorporate important ideas and techniques that would be implemented in modern ML. The data analytics tetrahedron proposed here can be applied to data of any size. To further contextualize, selected case studies capturing good practices are highlighted to provide pharmaceutical scientists, pharmaceutical ML enthusiasts, readers, reviewers, and regulatory authorities an exposure to fundamental and cuttingedge techniques of ML and data science with respect to chemistry, manufacture, and control (CMC) of drug products. In addition, the authors believe that leveraging ML within CMC procedures can assist in improving decision-making, increasing quality, and enhancing the speed of pharmaceutical product development. IOP Publishing Ltd 2023. All rights reserved. -
ML Use in Fraud Detection in the Financial Sector
Detecting fraud is complex, requiring analysis of vast data over time, making it resource- intensive for human investigators. This chapter explores key forms of financial fraud, including those involving corporations, banks, and insurance sectors. It provides a thorough evaluation of various machine learning techniques, such as supervised and unsupervised learning, deep learning, and ensemble methods, used to detect and prevent these financial crimes. The chapter examines the effectiveness of algorithms like LR, DT, SVM, and NN in identifying complex fraud patterns. Additionally, it discusses advanced approaches like anomaly detection, clustering, and real- time fraud detection systems. It highlights ongoing challenges and proposes potential future developments, including explainable AI, federated learning, and improved data processing techniques. This comprehensive exploration aims to deepen the understanding of machine learning's role in fraud detection and guide future research toward more effective and reliable fraud prevention strategies. 2025 by IGI Global Scientific Publishing. All rights reserved. -
ML-Based Fall Risk Prediction to Substitute Personal Assistance for Hospitalized Elderly: Integrating Geriatric Assessment and E-Health Records
Geriatric assessment serves as a holistic evaluation tool, encompassing various aspects of the elderly individual's health, including physical function, cognition, and psychosocial factors. Integration of CGA data with EHRs allows for a comprehensive analysis of the individual's health status and medical history, providing valuable insights into their risk factors for falls. The ML-based predictive model developed in this study utilizes these integrated data sources to identify patterns and trends associated with fall occurrences among hospitalized elderly patients. By analysing various variables, including mobility indicators, medication usage, and previous fall history, the model can generate accurate predictions of fall risks for individual patients. This ML-driven approach has the potential to significantly improve patient safety and quality of care by enabling healthcare providers to pre-emptively identify and address fall risks among hospitalized elderly individuals, thereby reducing the reliance on constant personal assistance while ensuring optimal patient outcomes. 2025 by IGI Global Scientific Publishing. -
ML-Based Prediction Model for Cardiovascular Disease
In this paper, the prediction of cardiovascular disease model based on the machine learning algorithm is implemented. In medical system applications, data mining and machine learning play an important role. Machine learning algorithms will predict heart disease or cardiovascular disease. Initially, online datasets are applied to preprocessing stage. Preprocessing stage will divide the data from baseline data. In the same way, CVD events are collected from data follow-ups. After that, data will be screened using the regression model. The regression model consists of logistic regression, support vector machine, nae Bayes, random forest, and K-nearest neighbors. Based on the techniques, the disease will be classified. Before classification, a testing procedure will be performed. At last from results, it can observe that accuracy, misclassification, and reliability will be increased in a very effective way. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
MLLR based speaker adaptation for indian accents
Speech Recognition has become an inherent and important feature of today's mobile based apps. Speech input is a very popular option for people with limitations of using the keyboard / mouse in a computer system. Nowadays, more voice messages are used than written text as they also convey the emotions of the speakers. As solutions are developed with native speakers of a language, many of the English input systems have higher accuracy for native speakers than for people with English as their second language (L2), especially for Asian population. The complexity increases since the accent and intonation of Indian speakers are varied from region to region and state to state. This paper analyses an effective speaker adaptation mechanism implemented with Indian speaker profiles and with a very small amount of adaptation data. This research is to facilitate a speaker adaptive system for the speech disabled users with limited disabilities like stuttering and/or unintelligible speech due to illness like cerebral palsy. Experimental results show improvements in the recognition accuracy for speakers speaking small sentences. 2017 University of Bahrain. All rights reserved. -
MMOF: A Multi-Metric Objective Function for Congestion Detection Under Varying Transmission Ranges in RPL-Based WSN
The Routing Protocol for Low Power Lossy Networks (RPL) is prone to congestion under high traffic. The single-path routing strategy and single-parent selection make RPL energy and resource-efficient only when the traffic is low and uniform. Two Objective Functions (OFs) are defined for RPL, which use single routing metrics-Expected Transmission Count (ETX) and hop count, to select the best parent and path toward the root. However, considering a single metric for OFs is unsuitable for detecting congestion in Lossy Networks (LLNs) applications as each metric has limitations. The current study proposes a novel Multi-Metric Objective Function (MMOF) that combines these two metrics and removes the weakness of the existing OFs. The proposed MMOF works under the nodes' varying transmission ranges (Tx ranges) to reduce the congestion. By changing Tx ranges, we show that the congestion in a fixed topology RPL network reduces, and MMOF can detect this congestion state more accurately than the existing OFs. The research introduces a successful transmission probability metric that makes MMOF more efficient in detecting congestion than ETX and Hop-Count. We prove that considering these two parameters individually is misleading and cannot contribute 100% to detect congestion state. Increasing transmission range can decrease congestion, and MMOF can detect this state transition with 100% accuracy. Simulation results in Cooja show that MMOF outperforms these two metrics and that the robust metric shows a linear relationship with the Tx range. Finally, two quality of service (QoS) parameters are derived to prove the method's efficiency and novelty. The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024. -
Mn2(CO)10 catalyzed visible-light-promoted synthesis of 1H-pyrazole-4-carboxamides; A sustainable multi-component statergy with antibacterial and cytotoxic evaluations
Multicomponent reactions play a pivotal role in synthesizing 1H-pyrazole-4-carboxamides, underscoring its significance in sustainable organic synthesis. These compounds, valued for their diverse biological activities, have garnered substantial attention in pharmaceutical research. A facile, rapid one-pot strategy to access an extensive array of 1H-pyrazole-4-carboxamide derivatives, utilizing substituted aldehydes, cyanoacetamide, and hydrazine hydrate as substrates and a readily accessible Mn2(CO)10 as photocatalyst in EL: H2O (1:1). Among the synthesized series, products 4b, 4 g, 4k showed remarkable antibacterial activity against E coli, P aeruginosa, S. aureus in agar medium and excellent cytotoxicity with Human colorectal carcinoma (HCT-116), Liver cancer cells (Hep-G2) and breast adenocarcinoma (MCF-7) cell lines. The current method is characterized by its affordability, non-toxicity, easy access to starting materials, and notably with minimal waste generation. Additionally, remarkable aspects include its mild operating conditions, environmentally friendly nature, and the ability to accommodate a wide range of both electron-donating and electron-withdrawing groups. 2024 The Author(s) -
MnO2 anchored NTi3C2 MXene as a bifunctional electrode for enhanced water splitting
The domain of energy research is vigorously exploring a wide array of materials, from advanced carbon-based substances like graphene and carbon nanotubes to emerging contenders like MXenes. Ti3C2 MXene offers exceptional performance in electrochemistry, benefiting from its remarkable electronic conductivity, considerable surface area, chemical stability, cost-effectiveness, hydrophilicity, and eco-friendliness. However, it undergoes self-accumulation, which diminishes the number of electrochemically active sites, resulting in decreased performance. In this study, MnO2 particles are intricately anchored onto the surfaces and within the layers of nitrogen-doped Ti3C2 (NTi3C2), resulting in the creation of innovative interface engineered NTi3C2/MnO2 nanosheets. Due to its distinctive heterostructure and favourable interfacial interaction, the NTi3C2/MnO2 electrode shows better performance in both the hydrogen and oxygen evolution reactions, exhibiting low overpotentials of 130 mV and 289 mV, respectively, at a current density of 10 mA cm?2. Furthermore, it requires a cell voltage of 1.7 V to achieve a current density of 10 mA cm?2 during the overall water splitting process. The NTi3C2/MnO2 composite also maintains sustained durability for a period of 4 h. This enhanced electrochemical activity of NTi3C2/MnO2 can be due to the synergistic effects resulting from the intricate contact between NTi3C2 and MnO2. This research presents a simple methodology for designing MXenes-based multicomponent electrodes for electrochemical water splitting reactions and its potential application for electrochemical water splitting. 2024 Hydrogen Energy Publications LLC -
MnO2 Nanoclusters Decorated on GrapheneModified Pencil Graphite Electrode for Non-Enzymatic Determination of Cholesterol
Electrochemically deposited MnO2 on graphene coated Pencil Graphite Electrode (PGE) has been used to develop a facile electrochemical sensor for the determination of Cholesterol. Cyclic voltammetric (CV) studies and electrochemical impedance spectroscopic (EIS) technique were used to investigate the electrochemical properties of the modified sensing platform. The physicochemical properties of the modified electrodes were characterized by X-ray photoelectron spectroscopy (XPS), Scanning electron microscopy (SEM), Transmission electron microscopy (TEM), X-ray diffraction (XRD) and Fourier transform infrared spectroscopy (FTIR). The experimental conditions such as effect of scan rate, concentration and pH were optimized. The linear dynamic range for the determination of Cholesterol was found to be 120?10 M2400?10 M under optimum conditions. The ultralow level of detection limit (0.42 nM) demonstrates the high sensitivity of the proposed method. The developed method was successfully applied for the non-enzymatic determination of Cholesterol in human blood samples at ultralow levels. 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim -
MnO2-Pi on biomass derived porous carbon for electro-catalytic oxidation of pyridyl carbinol
A facile electrochemical oxidation of pyridyl carbinol based on Manganese dioxide-Phosphate (MnO2-Pi) was fabricated by electro-deposition of MnO2-Pi on Porous carbon nanospheres (PCN) modified carbon fiber paper (CFP) electrode. A simple working electrode was developed initially by coating Monkey Pod (MP) derived PCN on carbon fiber paper (CFP) electrode. Voltammetric deposition of MnO2-Pi on PCN/CFP electrode was carried out in an electrolyte containing phosphate buffer and KMnO4. The modified electrodes (PCN/CFP and MnO2-Pi-PCN/CFP) were characterized by different physicochemical methods and electroanalytical techniques like cyclic voltammetry and AC impedance spectroscopy. Inorganic phosphate (Pi) and MnO2 centers present on PCN/CFP electrode plays a major role towards oxidation of pyridyl carbinol electrochemically. The proposed MnO2-Pi-PCN/CFP electrode was effectively applied for the electrochemical oxidation of pyridyl carbinol in TEMPO medium. 2020 The Author(s). -
Mobile Apps for Enhanced Bleisure Tourism Experiences: Exploring the Prospects and Challenges
Mobile applications play a pivotal role in enabling and enhancing bleisure travel experiences. These apps offer solutions for communication, itinerary planning, transportation booking, and leisure discovery, reflecting the evolving expectations of modern travelers for efficiency, flexibility, and customized experiences. Despite their benefits, challenges such as data privacy concerns and information overload persist. Looking ahead, the future of bleisure travel is poised for further transformation through advances in mobile technology, including augmented reality and artificial intelligence. However, a research gap exists in understanding the full spectrum of mobile apps catering to bleisure tourists' needs. This chapter aims to address this gap by classifying mobile apps for bleisure tourism, exploring their advantages, and identifying challenges and opportunities for innovation. By doing so, it seeks to contribute to a deeper understanding of the role of mobile technology in shaping the landscape of bleisure tourism in the digital age. 2024 by IGI Global. All rights reserved. -
Mobile apps in bleisure tourism: Enhancing travel experience, work-life balance, and destination exploration
This study aims to achieve four primary objectives: first, to evaluate how mobile apps improve travel productivity and efficiency by streamlining logistics and simplifying planning for both business and leisure activities; second, to investigate how these apps support the integration of work and leisure by providing tools for remote work, task management, and peer communication; third, to explore how mobile apps enhance the quality and authenticity of bleisure experiences by helping travelers discover new places and immerse themselves in local culture; and finally, to construct a comprehensive framework for mobile apps in bleisure tourism for use by multiple stakeholders, including travelers, travel companies, the hospitality industry, employers, local tourism boards, and app developers. This study highlights the significance of mobile technology in optimizing the bleisure travel experience. 2024 by IGI Global. All rights reserved. -
Mobile banking technology adaptation model: Revisiting the TAM approach /
Patent Number: 202041046897, Applicant: Dr.G Suresh.
The banking industry now is widely using the internet and mobile applications for rendering services to their customers. Customers now use the technology-enabled banking pervasively as it enables them to make self-controlled transactions€”the data collected through crowd-sourcing method, which considered to be the most relevant. The study found that perceived Trust on m-bankers has a direct effect on m-banking user behavioural intention. -
Mobile banking technology adoption model: Revisiting the tam approach
The user acceptance of Mobile banking technology is limited in terms of appropriate measurement variables. In M-banking practice, the influencing factors and the relationship with adoption is unrevealed. The data was collected through crowd sourcing method which is considered to be most relevant method. The results of hypotheses testing and SEM analysis showing that the relationship between the perceived usefulness and behavioral intention is significant direct relationship is existed in the same way perceived trust on the m-bankers has direct effect on m-banking user behavioral intention. The Perceived Ease of Use and Perceived Social Influence is not significantly influencing the behavioral intention but the important missed constructs, Perceived Trust along with Perceived Usefulness is highly influencing the M-banking users. The M-banking App developers should emphasis on need based apps and must incorporate strong security aspects for eliminating model risk associated with the M-banking application. The present study developed a new measurement variable called Perceived Social Influence and Perceived Trust along with Perceived Usefulness and Perceived Ease of Use of original TAM which are hypothesized to adopt M-banking technology. For M-Banking technology services, the original TAM did not hold good as there was an absence of a crucial factor for M-banking, Perceived Trust and Social Influence. 2019, Institute of Advanced Scientific Research, Inc. All rights reserved. -
Mobile banking technology adoption model: Revisiting the TAM approach /
Journal of Advanced Research In Dynamical And Control Systems, Vol.11, Issue 4, pp.1407-1415, ISSN No: 1943-023X. -
Mobile Freeze-Net with Attention-based Loss Function for Covid-19 Detection from an Imbalanced CXR Dataset
In this paper, we present a novel framework, that is, Mobile Freeze-Net along with Attention-based Loss Function, for Covid-19 detection from a Chest X-Ray (CXR) dataset. First, we have observed that by freezing 50% of a Mobile Net-V2 model (means fine-tuning 50% layers from ImageNet dataset) has automatically removed the class imbalance problem from the CXR dataset considerably. We call this 50% frozen Mobile Net-V2 model as Mobile Freeze-Net. Secondly, we have proposed an Attention-based Loss function, which provides more attention to the class, having higher inter-class similarity. We have computed attention weights for each class from the statistical inference of the dataset itself, by employing a Monte-Carlo method and thereafter, we have incorporated those weights into WCCE loss function of Mobile Freeze-Net model. By utilizing Mobile freeze-Net, we have achieved testing accuracy, F1 score, precision and recall of 93%, 94%, 93% and 94% respectively. This is approximately 3-4% improvement compared to 100% fine tuning of Mobile-Net V2. Furthermore, we have achieved approximate 1-2% improvement of Mobile Freeze-Net, after incorporating Attention-based Loss function. For the validity of the proposed framework, we have conducted experiments with 10-fold cross validation. All these experimental results suggest that our proposed framework has outperformed other existing models considerably. 2023 Owner/Author(s). -
Mobile in learning: Enhancement of information and communication technologies
The technological advancement in the world has changed the people's life. The people view point towards the usage of technologies in different fields like business, tourism, communication, education etc. has changed. Mobile learning can give flexible learning environ-ment for the user. It can also increase the participant number in the online teaching learning process. This paper discusses about the ef-fectiveness of the current technologies used in higher education system. It profiles the advantages of using mobile in accessing the uni-versity central system for teaching and learning. It also discusses about mobile digital book with augmentation, which can be used to improve the teaching and learning process of the different departments in the university. 2018 Authors. -
Mobile-Based Indian Currency Detection Model for the Visually Impaired
According to surveys held in 2019, India holds the largest population standing just after China, but when it comes to visually impaired people, India ranks number one. There are approximately 37 million people across India who are suffering from visual impairment. Special care and measures are taken to help these people live a peaceful life as any other citizen of India, but with the demonetization that happened in the recent years, the Indian economy was replaced with newer currency notes as an attempt to stop black money and fight corruption. Even though the objectives were clear and attainable, with the newer currency notes, the visually impaired people are facing various problems, as there is no provision for them to actually check the currency as the notes are not equipped with Braille system and the sizes of each and every currency is also the same in many cases. To counteract this problem, a mobile-based Indian currency detection model would be a better solution as it enables a visually impaired person to identify the value of specific currency he is holding. The mobile-based Indian currency detection model is the proposed model which will be using image processing for feature extraction and a basic CNN (convolutional neural network) for identification of currency with the given feature inputs. This model is being made into a mobile-based application so as to enable a visually impaired person to check for any possible frauds as fast as possible. 2020, Springer Nature Switzerland AG. -
MobileNetV3-Based Fine-Tuned Facial Emotion Recognition with Targeted Class Balancing
Facial emotion recognition (FER) is a pillar of affective computing and augmented human computer interaction, but has been stymied by the problem of class imbalance and lack of prevalence of subtle emotional differences. This paper presents a lightweight FER framework based on the MobileNetV3 architecture with a fine-tuned and weighted dataset that applies class balance and class weighting as strategies that optimized the three-class classification of three discrete emotions Angry, Happy, and Sad. The characteristics of the dataset were assembled comprising a total of 7,305 labelled facial images, based on the KDEF, Kaggle, and Face Expression Dataset hence inheriting the heterogeneity of subjects and imaging conditions. The pre-processing of all of the images carried out as the RGB input and after resizing (224 x 224 pixels) a massive data augmentation done to encourage generalization. Transfer learning in the training pipeline is done through progressive unfreezing and the weight of the loss on the minority classes (Angry and Sad) are boosted to improve the performance of detection. The achieved model resulted in an accuracy of 87% on the test set, and had equal accuracy in preciseness, recall, and F1-scores over all emotion types. Extended error analysis revealed that the majority of cases that were misclassified fell between the categories Angry and Sad because they were mistaken due to combining visual cues. Even then, the performance showed stability despite the variable lighting as well as in variable positional context. In Comparison, MobileNetV3 outperforms state-of-art-lightweight models with respect to accuracy and computation of similar computational complexity. 2025 IEEE.