Browse Items (16481 total)
Sort by:
-
A low cost and high actuation speed 3D printed prosthetic arm /
Patent Number: 202241047867, Applicant: Sujatha A K. -
A low cost and high actuation speed 3D printed prosthetic arm /
Patent Number: 202241047867, Applicant: Sujatha A K. -
A Low Voltage and Low Power Analog Multiplier
In this research work, a low voltage analog multiplier has been realized through the utilization of a flipped voltage follower (FVF). The multiplier is characterized by its capacity to function at low power while exhibiting high gain. The exclusive use of transistors in its implementation renders it highly appropriate for fully integrated circuit applications. The multiplier has been developed using a supply voltage of 500 mV and an operating frequency of 25 KHz. The design consumes power of 8.23 uW. Moreover, a comparative study between the proposed multiplier and the conventional gilbert multiplier is presented in the paper. All simulations and layout designs have been conducted through the virtuoso analog design environment (ADE) of Cadence at 45 nm CMOS technology. 2023 IEEE. -
A Low-Complexity Multiplier-Less Filter Bank Based on Modified IFIR for the SDR Channelizer
Digital filter banks are extensively used in an SDR channelizer for channelization. The objective of this research work is to design a low computational complexity FIR filter bank for generating sharp transition width channels for SDR. The design of unified and variable bandwidth channels for SDR using the proposed structure is based on the modified IFIR filter structure and cosine modulation technique (CMT). The performance of the proposed structure is demonstrated with the help of an example. The results show that the multiplier complexity of the proposed structure is less than those of other state-of-the art methods. The optimization techniques are incorporated in this work to further reduce the complexity of the proposed structure. With the help of canonical signed digit (CSD), multi-objective artificial bee colony (MOABC) and shift inclusive differential coefficients (SIDC) common sub-expression elimination (CSE) optimization, the filter used in this structure is made multiplier-less. 2024 IETE. -
A LSTM based model for stock price analysis and prediction
The share market in India is exceedingly unpredictable and volatile, with an infinite range of factors regulating the share market's orientations and tendencies; hence, forecasting the upswing and downturn is a difficult procedure. Because of several essential aspects, the principles of share market have always been unclear for shareholders. This study aims to significantly reduce the likelihood of analysis and forecasting with Long Short-term Memory (LSTM) model approach that is both resilient yet easy is still suggested. LSTM is a complete Learning Model that is a Predictive Method. Conversely, advancements in technology have opened the way for more efficient and precise share market forecasting in current times. Using the provided historical data sets, the results showed that the LSTM model has considerable potential for forecasting. 2023 Author(s). -
A Machine Learning Approach for Revving Up Revenue of Indian Tech Companies
This study addresses a critical gap in research by examining the effectiveness of various machine learning models in predicting revenue for Indian tech companies. The V.A.R, ARIMA, simple moving average, weighted moving average, and FB Prophet models were employed and their performances was compared. The findings demonstrate that FB Prophet consistently outperforms other models, exhibiting superior accuracy in revenue forecasting. This underscores FB Prophet's potential to offer precise revenue predictions, enabling companies to gain insights into their financial health, anticipate market trends, and optimize decision-making. Future research could further enhance accuracy by incorporating economic indicators, providing a more holistic view of revenue dynamics and empowering companies to make more informed strategic decisions. 2024 IEEE. -
A Machine Learning Approach to Consumer Behavior Analysis in Social Media-Influenced E-Book Markets
Social media has emerged as a dominant marketing channel, significantly influencing consumer purchase decisions. Despite extensive global research, little is known about region-specific dynamics in emerging markets such as India. This study addresses this gap by applying Random Forest and Gradient Boosting models to survey data from 386 respondents in the Delhi-NCR region to analyze e-book purchasing behavior. Data were preprocessed through encoding, normalization, and stratified traintest splitting (80:20), with reproducibility ensured via a fixed random seed. Model evaluation employed R, RMSE, and MAE metrics, alongside a paired-sample t-test. Results showed that Gradient Boosting (R = 0.82) outperformed Random Forest (R = 0.78; p = 0.038). Feature importance analysis revealed that behavioral variablespurchase intention, brand awareness, and social media engagementwere the strongest predictors, whereas demographic features contributed minimally. These findings emphasize the primacy of behavioral traits in social mediadriven e-book markets and provide evidence for designing region-specific digital marketing strategies in emerging economies. 2025, Interdisciplinary Publishing Academia. All rights reserved. -
A Machine Learning Approach to Crude Oil Price Prediction Using Support Vector Machine (SVM)
Crude oil is one of the most important energy sources, and fluctuations in its international prices affect all aspects of the economy. The price of crude oil is influenced by several variables, and the length of time that each component has an effect differs giving an increase in non-linear oil price features. Although it is a complex task, identifying the most essential factor influencing for precise predicting, crude oil prices are essential. Therefore, this study aims to employ a machine learning model to address the intricate relationships among different factors. Primarily, it gathers data regarding West Texas Intermediate (WTI) and Brent crude oil prices as well as macroeconomic variables. Secondly, the data is normalized to prepare it for further analysis. Finally, a crude oil prediction model is constructed using Support Vector Machine (SVM) to predict future international crude oil prices. The daily, weekly, and monthly prices are used to confirm the models efficacy developed using WTI and Brent oil. The models performance is also evaluated by incorporating various combinations of macroeconomic variables to find the most influential factor. Results from experiments show indicates the benchmark model was much exceeded by the developed model and performed very well in terms of prediction accuracy. The findings reveal that selecting the appropriate variables can greatly enhance prediction accuracy. This model has the potential to provide valuable insights for traders, investors, and energy-related enterprises, offering beneficial guidance for decision-making purposes. The Author(s), under exclusive license to Springer Nature Switzerland AG 2026. -
A Machine Learning Approach to Modelling Sales Performance Based on Social Media Analysis
By address the strategic challenges faced while maximising the advertising effectiveness, this paper examines the predictive relationship between different advertising sales platform and the brand sales performance. Four different supervised machine learning models - Random Forest, XGBoost, LightGBM, and a voting classifier ensemble model are applied to categorise the brand sales performance. The models are trained on a multi-platform advertising dataset. This research focuses on both the accuracy and interpretability of the data to make it easier identifying which platform affects the brand sales the most, which is quite different from the existing approaches that concentrates on finding predictive accuracy. This study aims to help marketers and businesses to make better and well informed data-driven decisions for marketing strategies by finding the most effective advertising platform. Each of the models were evaluated using Classification metrics which includes Accuracy, Precision, Recall and F1 score, along with confusion matrix. Future scope of the paper includes merging the models in real time systems and also by expanding it to carefully examine different time periods and customer groups. The Author(s), under exclusive license to Springer Nature Switzerland AG 2026. -
A Machine Learning Entrenched Brain Tumor Recognition Framework
Brain tumor detection plays a significant role in medical image processing. Treatment for patients with brain tumors is primarily dependent on faster detection of these tumors. More rapid detection of brain tumors will help in the improvement of the patient's life chances. Diagnosis of brain tumors by doctors most commonly follow manual segmentation, which is difficult and time-consuming; instead, automatic detection is necessary. Nowadays, automatic detection plays a vital role and can be a solution to detecting brain tumors with better performance. Brain tumor detection using the MRI images method is an essential diagnostic tool for predicting brain tumors; the implementation for these kinds of detection can be done using various machine learning algorithms and methodologies. It helps the doctors understand the actual progression of the evolving tumor, allowing the doctors to decide how the treatment has to be given for that particular patient and measures required to follow up. Therefore, the intention is to create a framework to detect brain tumors in MRI images using a machine learning algorithm and analyze the performance of the brain tumor detection using sensitivity and specificity, which helps us to analyze how well the algorithm has performed in detecting the brain tumors accurately and develop a mobile application framework in which the MRI images can be directly scanned to know whether the cancer is present in a scanned MRI image or not. 2022 IEEE. -
A Machine Learning Model for Augmenting the Media Accessibility for the Disabled People
In an era characterized by the proliferation of digital media, the need to efficiently use multimedia content has become paramount. This article discusses an innovative technique called 'Fast Captioning (FC)' to improve media accessibility, especially for people with disabilities and others with time restrictions. Modern Machine Learning (ML) algorithms are incorporated into the framework, which speeds up video consumption while maintaining content coherence. The procedure includes extracting complex features like Word2Vec embeddings, part-of-speech tags, named entities, and syntactic relationships. Using annotated data, a ML model is trained to forecast semantic similarity scores between words and frames. The predicted scores seamlessly integrate into equations that calculate similarity, thus enhancing content comprehension. Through this all-encompassing approach, the article offers a comprehensive solution that balances the requirements of contemporary media with the accessibility requirements of people with disabilities, producing a more inclusive digital environment. Machine Learning-based Media Augmentation (ML-MA) has achieved the highest accuracy of 96%, and the captioning is accurate. 2023 IEEE. -
A machine learning model for population analysis among different states in India which influences the socio, demographic and economic needs of society
In this work Data from 2011 census is taken to identify the state which influences more in Population census among the different states identified. The data is considered from Madhya Pradesh, followed with Utter Pradesh, then to Bihar, Bengal and Orissa. Similarly other case studies are also done for Southern Indian states and North Eastern States. Genetic algorithm will be tried to find the optimal location for the given study. A fitting function is calculated for the population data of 2011 using Lagrange Interpolation technique. This fitting function is given as input to Genetic algorithm to find the optimal state which have maximum influence in the population growth among different states of India as per the Case studies done. BEIESP. -
A machine learning model to predict suicidal tendencies in students
[No abstract available] -
A Machine Learning- Based Driving Assistance System for Lane and Drowsiness Monitoring
Lane line detection is a vital component when driving heavy vehicles; this concept follows the path for driving a vehicle to prevent the risk of accidentally entering another lane without the drivers knowledge, which could result in an accident. To detect the lane, use frame masking and Hough line transformation with efficient machine learning algorithms, pre-processed and trained adequately for optimum accuracy as per the provided dataset to spot the white markings on both sides of the lane. Long-distance truck drivers suffer from sleep deprivation, making driving extremely dangerous while tired and they ignore the line markings and wander into the wrong lane. This chapter proposes a portable system that does not require any sensors or interference with the vehicles wiring system; instead, a system that fits on a windshield or any surface to monitor the actions of the driver, using computer vision and feature-extracted datasets within a trained neural network model using cameras. This driver-assisted system can detect drowsiness and give an alarm to wake up the driver by identifying the Region of Interest. These predictions are made based on eye movements, and the algorithm generates a score. The higher the score, the longer the time between alarms. 2024 Taylor & Francis Group, LLC. -
A Machine Learning-Based Cross-Layer DoS Attack Detection Technique for IoT
The most potent and common attacks on Internet of Things (IoT) are Denial of Service (DoS) attacks. Unfortunately, because the attack occurs on numerous layers, a single layer detection method is insufficient and ineffective to counteract these attacks. The current work focuses on the detection of cross-layer DoS assaults using Machine Learning-based multiclass classifiers. Three attacks against Routing Protocol (RPL) and Transmission Control Protocol (TCP) are detected using three ML Classifiers (KNN, Gradient Boosting, Random Forest). The novelty of the study is the design of cross-layer attack datasets using feature engineering technique. The performances of the classifiers are analyzed in presence of both balanced and imbalanced datasets. The results show that Gradient Boosting classifier has highest accuracy of up to 98% with deviation of up to 97%. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
A Malicious Botnet Traffic Detection Using Machine Learning
Detection of incorrect and malign data transfers in the Internet of Things (IoT) network is important for IoT safety to observe an eye on and prevent unwelcomed traffic flow to the network of IoT. For it, Machine Learning (ML) strategic methods are produced by several researchers to prevent malign data flows through the network of IoT. Nonetheless, because of the wrong choice of feature, a few malign Machine Learning models differentiate especially the movement of malign traffic. Still, what matters is the problem that needs to be deliberated in-depth to select the best features for better malign traffic acquisition in the network of IoT. Dealing with the challenge, a new process was proposed. 1st, the metric method of selecting a novel feature called the proposed CorrAUC, and hinged on CorrAUC, a new highlight for choosing the Corrauc algorithm name is also being developed, designed hinged on the system folding filter features precisely and select the active features of the choose ML method using AUC metric. After that, we apply a combined application Order of Preference by Similarity to Ideal Solution Using Shannon Entropy (TOPSIS) built on a bijective set which is soft to verify selected features for identification of malign 1traffic in IoT network. We test our method using data set of Bot-IoT and 4 dissimilar ML classifiers. Practical outcomeanalysis showed that our proposed approach works as well and can achieve greater than 96% results on average. 2022 Wolters Kluwer Medknow Publications. All rights reserved. -
A Markovian risk model with possible by-claims and dividend barrier
A MAP/PH risk model with possible by-claims and a dividend barrier is considered. Along with the main claim, a by-claim also can occur with a certain probability but by-claims are settled only after an inquiry and hence delayed. The model is analysed considering associated Markovian fluid models under the original timeline and an auxiliary timeline. Systems of integro differential equations (IDE) are developed for the Gerber-Shiu function (GSF) and the total dividends paid until ruin. Explicit expressions are obtained for the GSF of the models without and then with the barrier. Expressions are also provided for the moments of the total dividends paid until ruin. A dividends-penalty identity is given. The method is numerically illustrated with a two-phase model and sensitivity analysis of the model is done by varying some of the parameters involved. 2023 Inderscience Enterprises Ltd.. All rights reserved. -
A mathematical approach to the study on alkylating agents
There are several classes of anticancer drugs, among which our study focuses on alkylating agents. As a chemical graph invariant number, topological index, has crucial role in predicting the physical, chemical, biological and toxicity properties of a molecule. Different versions of Zagreb indices correlate well with various physio-chemical properties of a molecule. We are analysing physio-chemical properties of the class of alkylating agents using various Zagreb indices. In this paper we are able to predict the physico-chemical properties of a molecule which is not yet discovered using the Zagreb class. 2022 Author(s). -
A Mathematical Correlation of Compressive Strength Among Silica, Alumina and Calcia Present in Composite Red Mud and Iron Ore Tailingbricks
Waste Red Mud generated from bauxite beneficiation in aluminium industry contains sodium oxide in minor amount along with silica and alumina in significant quantities. Waste iron ore tailings from beneficiation of iron ore in steel industry contain silica and alumina in significant quantities. A combination of both these materials in different amounts along with GGBS and lime addition resulted in complex alkali-activated reaction products consisting of (Si/Al), (Ca/Si) and (Ca/(Si+Al)) complexes which influence compressive strength of the test samples on curing for extended time periods at room temperature. Individual correlation coefficients of these complexes with compressive strength yielded high values with (Si/Al) and (Ca/Si+Al) ratios (0.92 and 0.96, respectively) while showing a poor correlation coefficient with (Ca/Si) ratio (0.88). A direct regression analysis between compressive strength and (Si/Al) ratio and (Ca/Si+Al) ratio indicated negative values with (Si/Al) ratios but positive values with (Ca/ (Si+Al)) ratios. It is therefore concluded that the addition of lime and GGBS (contributed from both GGBS and lime addition) resulted in Ca-Si-Al complex formations which are responsible for improved compressive strength of the samples. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
A mathematical model that describes the relation of low-density lipoprotein and oxygen concentrations in a stenosed artery
The cellular activities of the endothelium layer between lumen and intima are significantly linked to the origin of the disease atherosclerosis. Three stages of atherosclerosis were investigated in this study (40%-mild, 50%-modest, and 60%-acute) concerning the coronary arterial segment. The essence of the hemodynamic factors like flow velocity, pressure, and wall shear stress has been investigated, as well as the interrelationships between them. At all degrees of stenosis, the biophysical relationship between convection-diffusion of low-density lipoproteins (LDL) and convection-diffusion of oxygen in the bloodstream is investigated. The Finite Element Methods (FEM) are used to solve the modeled partial differential equation systems. The method adopted is numerical in nature providing accurate graphical solutions to the framed systems. The physical effects of the deposition of LDL on the arterial wall, like a decrease in the diameter of the lumen, and toughening of the walls, are analyzed through the evaluation of the physical parameters. The study revealed that the deposition of LDL molecules in the post stenotic region leads to the depletion of oxygen in the region leading to the rapid dysfunctioning of the endothelial layer of the lumen-intima boundary. 2022 World Scientific Publishing Company.