Dr MANJAIAH D H
@mangaloreuniversity.ac.in
PROFESSOR
MANGALORE UNIVERSITY
RESEARCH INTERESTS
o Narrativeattentiveness of Routing Algorithms for Mobile Ad - Hoc Networks
o Explore challenging intend characteristics of 4G, 5G, WSN and VAN
o Significance of Cyber Security and Digital Forensics at current arena
o Predictive Data Analytics using Machine Learning and Deep Learning
o Effectiveness of Load Balancing and Virtualization in Cloud Computing Environment
Scopus Publications
Scopus Publications
Mohammad Kazim Hooshmand, Manjaiah Doddaghatta Huchaiah, Ahmad Reda Alzighaibi, Hasan Hashim, El-Sayed Atlam, and Ibrahim Gad
Elsevier BV
Amrithkala M. Shetty, Mohammed Fadhel Aljunid, D. H. Manjaiah, and Ahammed M. S. Shaik Afzal
Springer Nature Singapore
Amrithkala M. Shetty, Mohammed Fadhel Aljunid, and D. H. Manjaiah
Springer Nature Singapore
Amrithkala M. Shetty, Mohammed Fadhel Aljunid, and D. H Manjaiah
Informa UK Limited
Online reviews play a significant role in the success of a business. Deep learning models have emerged as crucial tools in this domain, with one-dimensional Convolutional Neural Network (1D CNN) being commonly used. However, this paper proposes a novel approach utilizing a Two-Dimensional Convolutional Neural Network (Att + 2D CNN) with attention mechanism, which effectively captures the dimensionality of the input text, resembling a 2D matrix. To further enhance the model’s performance, we employ pretrained word embeddings, specifically GloVe and Word2Vec. We thoroughly analyze the performance of these embeddings in conjunction with deep learning models. Remarkably, our proposed method, leveraging 2D CNN with attention, consistently achieves superior accuracy when compared to other models, specifically on Amazon Cell Phone reviews and Amazon Kindle reviews datasets, for both balanced and unbalanced natures. By employing this novel methodology, we demonstrate the ability to extract valuable insights from online reviews, enabling businesses to make informed decisions.
D. H. Manjaiah, M. K. Praveena Kumari, K. S. Harishkumar, and Vivek Bongale
Springer Nature Singapore
Shubha R. Shetty and D. H. Manjaiah
Springer Nature Singapore
Vedavati Bhandari and Manjaiah D.H
Informa UK Limited
Vedavati Bhandari and Manjaiah Doddaghatta Huchaiah
Springer Science and Business Media LLC
Mohammed Fadhel Aljunid and Manjaiah Doddaghatta Huchaiah
Elsevier BV
T. R. Tejasvi and D. H. Manjaiah
Wiley
Shubha R Shetty and Manjaiah D H
IEEE
Vehicles are imparting messages with other vehicles in ad hoc network, which is referred as VANET communication. Due to the emerging of new technologies as part of industry 4.0 the VANET as frequently used in communication applications. But in this cyber attacks are increased in the VANETS communications. Scholars have proposed different solutions and algorithms to find the attacks. Here we have proposed a method that uses artificial intelligence. The proposed method is the combination of Neural Network based Multilayer Perceptron (MLP) trained Fuzzy Logic System procedure to spot unusual behavior of automobiles in the ad hoc network. To validate the results time of detection, positive rate, and ratio of detection is used. The outcome will giv the better performance existing methods.
Suparna N and Manjaiah DH
IEEE
The owners of state-of-the-art residential villas always thrive for feature-rich connected devices to secure personal dwellings to ensure that they are always safely connected with house surroundings and appliances. Despite the luxury of connectedness, smart home devices lead to the risk of data security and several types of security attacks. The advent of IoT has revolutionized the way smart homes function and are secured.[1]. Vulnerable local networks, weak IoT devices and unencrypted data communication between the sensors pose threats to SHNs. Though the various levels of protocols provide necessary security features for IoT -based home networks, the data communication path between the sensors and the gateways at the smart homes is vulnerable and is open to various types of attacks. Therefore, in this paper, we successfully implemented Light Weight Cryptography Algorithm IDEA in Smart Home Networks and analysed the performance of the network.
John W. Kasubi and Manjaiah D. Huchaiah
IEEE
Feature selection plays an essential role in machine learning for reducing irrelevant, noisy, and redundant features, and selecting the optimal set of attributes for creating compelling predictive models of the study dataset subject. The feature selection (FS) methods entail deciding which essential features to use in machine learning (ML) for model development, and removing redundant features which decrease the overall classification accuracy. This study focused on a comparative analysis of three popular FS approaches: filter, wrapper, and embedded. In this regard, we selected a single technique from each approach, filter-based (Information Gain (IG)), wrapper–based (Recursive Feature Elimination (RFE)), and embedded–based (Tree-based) feature selections. Thereafter, the study applied five base learner classifiers: Logistic Regression (LR), Linear Discriminant Analysis (LDA), Naïve Bayes (NB), Random Forest (RF), and K-nearest neighbor (KNN), to build the human activity recognition (HAR) model. The experimental results show that RF outscored compared to the other classifiers when applied under tree-base feature selection, with an accuracy of 98.9% in house A, and 99.8% in house B. The FS methods enhanced forecast accuracy in the ARAS dataset more than before FS.
Ezz El-Din Hemdan and D. H. Manjaiah
Springer International Publishing
John W. Kasubi and D. H. Manjaiah
Springer Singapore
Shubha R. Shetty and D. H. Manjaiah
Springer Singapore
Daniel Mesafint Belete and Manjaiah D. Huchaiah
Informa UK Limited
Mohammed Fadhel Aljunid and Manjaiah Doddaghatta Huchaiah
Institution of Engineering and Technology (IET)
Manjaiah Doddaghatta Huchaiah and John William Kasubi
Springer Science and Business Media LLC
Ezz El-Din Hemdan and D.H Manjaiah
Springer Science and Business Media LLC
John W. Kasubi and Manjaiah D. Huchaiah
Springer International Publishing
Daniel Mesafint Belete and Manjaiah D. Huchaiah
Springer Singapore
Mani Bushan Dsouza and D. H. Manjaiah
Springer Singapore
K. Ankitha and D. H. Manjaiah
Springer Singapore
The clinical mastitis is a harmful disease in cows and many researchers working on milk parameters to detect clinical mastitis. Internet of things (IoT) is a developing era of technology where every object is connected to the Internet using sensors. Sensors are an essential unit of an IoT to collect the data for analysis. The proposed method concentrates on deploying sensors on cows to monitor the health issues and will state IoT as an Internet of Animal Health Things (IoAHT). Dairy cows are an essential unit of the Indian economy because India is a top country in milk production. Clinical mastitis affects dairy cows in the production of milk. Recent studies in the dairy industry proved the use of technologies and sensors for good growth of cows. This paper reviews a method used for detecting clinical mastitis in cows and proposes a system for the same using IoAHT. The KNN and SVM algorithms are used on the primary data set to obtain a result of the detection. In comparison to these algorithms, SVM provided better results in detecting mastitis in cows.
Mohammed Fadhel Aljunid and Manjaiah Doddaghatta Huchaiah
Institution of Engineering and Technology (IET)
As a result of a huge volume of implicit feedback such as browsing and clicks, many researchers are involving in designing recommender systems (RSs) based on implicit feedback. Though implicit feedback is too challenging, it is highly applicable to use in building recommendation systems. Conventional collaborative filtering techniques such as matrix decomposition, which consider user preferences as a linear combination of user and item latent features, have limited learning capacities, hence suffer from a cold start and data sparsity problems. To tackle these problems, the research direction towards considering the integration of conventional collaborative filtering with deep neural networks to maps user and item features. Conversely, the scalability and the sparsity of the data affect the performance of the methods and limit the worthiness of the results of the recommendations. Therefore, the authors proposed a multi-model deep learning (MMDL) approach by integrating user and item functions to construct a hybrid RS and significant improvement. The MMDL approach combines deep autoencoder with a one-dimensional convolution neural network model that learns user and item features to predict user preferences. From detail experimentation on two real-world datasets, the proposed work exhibits substantial performance when compared to the existing methods.