Vellore Institute of Technology (VIT) is a private university located in&nbsp;Tamil Nadu, India. Founded in 1984, as Vellore Engineering College, the institution offers 20 undergraduate, 34 postgraduate, four integrated and four research programs. It has campuses in Vellore, Amravati, Bhopal and Chennai.

VIT is one of the top ranked private universities in India according to NIRF, THE and QS Rankings.&nbsp;Govt. of India has recognized&nbsp;VIT, Vellore as an&nbsp;Institution of Eminence. This has allowed VIT to take independent quality initiatives and move up in world ranking.

&nbsp;

&nbsp;

VIT University

Department of Information Technology

The development in technology and science has contributed to a vast volume of data from various agrarian fields to be aggregated in the public domain. Predicting the crop yield based on climate, soil and water parameters has been a potential re- search subject. Therefore an objective arises in integrating the data with agrarian processes for crop enhancement. In this paper, a new hybrid regression-based algorithm, Reinforcement Random Forest is proposed which displays significantly enhanced performance over traditional machine learning techniques like the random forest, decision tree, gradient boosting, artificial neural network and deep Q-learning. The new strategy executes reinforcement learning at every selection of a splitting attribute amid the process of tree construction for the efficient utilization of the available samples. They analyze the variable significance measure to select the most substantial variable for node splitting process in the model development and promotes efficient utilization of training data. This integrated hybrid procedure provides significant enhancement over prevailing strategies specifically for sparse model structures. Besides performing internal cross-validation, the proposed approach requires less parameter tuning, reduces over-fitting, faster calculation and more transparent. The experimented models are evaluated with various assessment metrics like Root Mean Squared Error, Mean Squared Error, Determination Coefficient and Mean Absolute Error. The results obtained delineated that the proposed approach performs better with reduced error measures and improved accuracy of 92.2%. © 2021, Springer-Verlag GmbH Germany, part of Springer Nature.

World University root node

School of Information Technology and Engineering

Vellore Campus

Journal of Ambient Intelligence and Humanized Computing

A reinforced random forest model for enhanced crop yield prediction by integrating agrarian parameters

Semantic similarity, sometimes referred as semantic relatedness, is one of the important concepts that help in various applications that involve natural language processing. In literature, there are plenty of similarity measures to compute the relationship among words in monolingual and cross-lingual documents. They help us in understanding text, finding plagiarism, information retrieval etc. They can be categorised based on the resources used into corpus-based and knowledge-based measures. These measures are plenty for the English language. For the Tamil language, there are hardly any works in calculating the similarity between words. In this paper, we proposed a similarity finding technique that exploits the knowledge from the resources like Tamil Indo WordNet, Tamil Wikitionary and Oxford Tamil Dictionary. We have used the definitions and example sentences of each word that are available through each of these resources for similarity calculation. The proposed approach is evaluated using human evaluated Miller Charles and Rubenstein Goodenough datasets. Copyright © 2021 Inderscience Enterprises Ltd.

International Journal of Innovative Computing and Applications

Hybrid approach for semantic similarity calculation between Tamil words

Hyperspectral imagery is one of the research area in the field of Remote sensing. Hyperspectral sensors record reflectance (also called spectra signature) of object or material or region across the electromagnetic spectrum. Mineral identification is an urban application in the field of Remote sensing of Hyperspectral data. EO-1 hyperion dataset is unlabeled data. Various types of clustering algorithms are proposed to identify minerals. In this work principal component analysis is used to reduced it’s dimension by reducing bands. Hard-clustering and soft-clustering algorithms are applied on given data to classify the minerals into classes. K-means is hard type of clustering which classify only non-overlapping cluster however, PFCM is soft type of clustering which allow a data points to belongs more than one cluster. Further, results are compared using cluster validity index using DBI value. Both clustering algorithms are experiments on original HSI image and reduced bands. Result shows that PFCM is perform better than K-means for the both type of images. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.

Department of Analytics

School of Computer Science and Engineering

Advances in Intelligent Systems and Computing

Unsupervised learning method for mineral identification from hyperspectral data

Botnet attacks are the new threat in the world of cyber security. In the last few years with the rapid growth of IoT based Technology and networking systems connecting large number of devices, attackers can deploy bots on the network and perform large scale cyber-attacks which can affect anything from millions of personal computers to large scale organizations. Hence, there is a necessity to implement countermeasures to over-come botnet attacks. In this paper, three hybrid models are proposed which are developed by integrating multiple machine learning algorithms like Random Forest (RF), Support Vector Machine (SVM), Naive Bayes (NB), K-Nearest Neighbor (KNN) and Linear Regression (LR). According to our experimental analysis, the RF-SVM has the highest accuracy (85.34%) followed by RF-NB-K-NN (83.36%) and RF-KNN-LR (79.56%). © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.

Department of Software and Systems Engineering

Identification of botnet attacks using hybrid machine learning models

Semantic similarity quantification between concepts is one of the inevitable parts in domains like Natural Language Processing, Information Retrieval, Question Answering, etc. to understand the text and their relationships better. Last few decades, many measures have been proposed by incorporating various corpus-based and knowledge-based resources. WordNet and Wikipedia are two of the Knowledge-based resources. The contribution of WordNet in the above said domain is enormous due to its richness in defining a word and all of its relationship with others. In this paper, we proposed an approach to quantify the similarity between concepts that exploits the synsets and the gloss definitions of different concepts using WordNet. Our method considers the gloss definitions, contextual words that are helping in defining a word, synsets of contextual word and the confidence of occurrence of a word in other word's definition for calculating the similarity. The evaluation based on different gold standard benchmark datasets shows the efficiency of our system in comparison with other existing taxonomical and definitional measures. © 2021 World Scientific Publishing Co.

Journal of Information and Knowledge Management

LIS4: Lesk Inspired Sense Specific Semantic Similarity using WordNet

This study focuses on creating offline benchmarks for handwritten characters in the local Tamil Brahmi language in India. This study has generated corpus of 1 lakh and 92000 character of brahmi script by getting offline handwritten scripts from 1600 writters. Hough transform and contour techniques are used to segment and extract scripts from the sheet. The offline data is also stored in csv format. The database is, however, intended to be useful for the identification of linguistic tasks. It will encourage the research work on Tamil Brahmi script handwriting recognition by providing free access to researchers. The designed dataset consists of 9 vowels, 18 consonants, and 209 classes so that the researchers can use machine learning algorithms to perform an extended research analysis. Handwritten data is collected and applied to create any linguistic frameworks for script analyzer. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021.

Lecture Notes on Data Engineering and Communications Technologies

TAMIZHİ: Historical Tamil Brahmi Handwritten Dataset

Training a Machine Learning (ML) model on bigger datasets is a difficult task to accomplish, especially when a high-end configuration is not accessible. A relatively good configuration may also not always produce quick outcomes and depending on the dataset size, the time taken would be anything between seconds to several hours. More often, the tasks we are interested in involve big datasets and complex models. The purpose of our work was to see how effective Hadoop can be in terms of increasing the efficiency of working with Machine Learning for a given problem. Out of many models to choose from, Logistic Regression was chosen, which is relatively simpler to implement. Three Logistic Regression models were implemented and trained on MNIST Handwritten Digits dataset. First one was implemented in Python using NumPy without any ML libraries. The second implementation used LogisticRegression class that comes with the Scikit-learn Python package, and the third implementation was done using PySpark MLlib. Towards the end of the paper, we present the observations and results obtained from the execution of each. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.

Logistic regression on hadoop using pyspark

To curb the accident rate and traffic levels, strict implementation of the rules and continuous monitoring of the traffic is mandatory. Traffic Rule Violation Monitoring System ensures that the rules are followed strictly and it reduces the human effort. The main objective of this work is to identify the Triple Riding. To detect the triple riders, the deep learning framework darknet is used, which in turn uses a type of convolutional neural networks i.e. Deconvolutional neural network-based YOLO (You Only Look Once) algorithm for detection of the number of persons riding a bike, the system classifies the vehicle as to the rule-breach vehicle or not. The junctions acting as the data collections center, collects the data. The image of the vehicle classified as the rule-breach is stored along with the data such as vehicle manufacturing ID and vehicle speed transferred at the particular frame. The transfer of the data is facilitated using the GSM module and the NodeMCU deployed on the vehicle. The vehicle number will be verified with the transport office. To survive the lack of internet connectivity or low internet connectivity, the system is being equipped with the GSM module; else, the data related to the vehicle can be pulled by the development boards deployed at the junctions, acting them as the central part of the public internetwork deployed. This public internetwork acting the medium to pull the data from the vehicle to the central system. This is carried out using the concept of dynamic network configuration in NodeMCU. The use of Node MCU and the public network system makes the system much more viable, available and reliable. Thereby making the riders follow the rules properly and reducing irresponsible driving. © 2020, Springer Science+Business Media, LLC, part of Springer Nature.

Multimedia Tools and Applications

Detection of the triple riding and speed violation on two-wheelers using deep learning algorithms

Word sense disambiguation is the way to compute the correct sense of a word. It is considered as one of the important subtasks in natural language processing, machine translation and information retrieval. WSD found improving the overall performances of these systems. The job of WSD is to eliminate all senses of a word except the appropriate one as per the given context. The work in Tamil linguistics domain for information retrieval or natural language processing is very less. WSD can be performed in supervised and unsupervised manner. Here, we have proposed an unsupervised approach to disambiguate Tamil words in a given context using the context words and their dictionary gloss definitions. We have proposed two variants of our approach. The first approach uses the number of word overlapping between the glosses of context words whereas the second one uses the similarity between the glosses of context words with that of the ambiguous word. The second one found best among the two. For our approach, we have used Tamil Indo-WordNet, Oxford Tamil Dictionary and English WordNet dictionary glosses. Our method achieves better result in recognising correct senses in Tamil text. Copyright © 2021 Inderscience Enterprises Ltd.

International Journal of Intelligent Enterprise

Word sense disambiguation in Tamil using indo-wordnet and cross-language semantic similarity

Facial expression is the crucial component for human beings to express their mental state and it has become one of the prominent areas of research in computer vision. However, the task becomes challenging when the given facial image is non-frontal. The influence of poses on facial images is alleviated using an encoder of a generative adversarial network capable of learning pose invariant representations. State-of-art results for image generation are achieved using styleGAN architecture. An efficient model is proposed to embed the given image into the latent vector space of styleGAN. The encoder extracts high-level features of the facial image and encodes them into the latent space. Rigorous analysis of semantics hidden in the latent space of styleGAN is performed. Based on the analysis, the facial image is synthesized, and facial expressions are recognized using an expression recognition neural network. The original image is recovered from the features encoded in the latent space. Semantic editing operations like face rotation, style transfer, face aging, image morphing and expression transfer can be performed on the image obtained from the image generated using the features encoded latent space of styleGAN. L2 feature-wise loss is applied to warrant the quality of the rebuilt image. The facial image is then fed into the attribute classifier to extract highlevel features, and the features are concatenated to perform facial expression classification. Evaluations are performed on the generated results to demonstrate that state-of-art results are achieved using the proposed method. © 2021 the Author(s), licensee AIMS Press.