Shubham Bharti, Arun Kumar Yadav, Mohit Kumar and Divakar Yadav
With the rise of social media platforms, an increasing number of cases of cyberbullying has reemerged. Every day, large number of people, especially teenagers, become the victim…
Abstract
Purpose
With the rise of social media platforms, an increasing number of cases of cyberbullying has reemerged. Every day, large number of people, especially teenagers, become the victim of cyber abuse. A cyberbullied person can have a long-lasting impact on his mind. Due to it, the victim may develop social anxiety, engage in self-harm, go into depression or in the extreme cases, it may lead to suicide. This paper aims to evaluate various techniques to automatically detect cyberbullying from tweets by using machine learning and deep learning approaches.
Design/methodology/approach
The authors applied machine learning algorithms approach and after analyzing the experimental results, the authors postulated that deep learning algorithms perform better for the task. Word-embedding techniques were used for word representation for our model training. Pre-trained embedding GloVe was used to generate word embedding. Different versions of GloVe were used and their performance was compared. Bi-directional long short-term memory (BLSTM) was used for classification.
Findings
The dataset contains 35,787 labeled tweets. The GloVe840 word embedding technique along with BLSTM provided the best results on the dataset with an accuracy, precision and F1 measure of 92.60%, 96.60% and 94.20%, respectively.
Research limitations/implications
If a word is not present in pre-trained embedding (GloVe), it may be given a random vector representation that may not correspond to the actual meaning of the word. It means that if a word is out of vocabulary (OOV) then it may not be represented suitably which can affect the detection of cyberbullying tweets. The problem may be rectified through the use of character level embedding of words.
Practical implications
The findings of the work may inspire entrepreneurs to leverage the proposed approach to build deployable systems to detect cyberbullying in different contexts such as workplace, school, etc and may also draw the attention of lawmakers and policymakers to create systemic tools to tackle the ills of cyberbullying.
Social implications
Cyberbullying, if effectively detected may save the victims from various psychological problems which, in turn, may lead society to a healthier and more productive life.
Originality/value
The proposed method produced results that outperform the state-of-the-art approaches in detecting cyberbullying from tweets. It uses a large dataset, created by intelligently merging two publicly available datasets. Further, a comprehensive evaluation of the proposed methodology has been presented.
Details
Keywords
Adarsh Kumar, Saurabh Jain and Divakar Yadav
Simulation-based optimization is a decision-making tool for identifying an optimal design of a system. Here, optimal design means a smart system with sensing, computing and…
Abstract
Purpose
Simulation-based optimization is a decision-making tool for identifying an optimal design of a system. Here, optimal design means a smart system with sensing, computing and control capabilities with improved efficiency. As compared to testing the physical prototype, computer-based simulation provides much cheaper, faster and lesser time-and resource-consuming solutions. In this work, a comparative analysis of heuristic simulation optimization methods (genetic algorithms, evolutionary strategies, simulated annealing, tabu search and simplex search) is performed.
Design/methodology/approach
In this work, a comparative analysis of heuristic simulation optimization methods (genertic algorithms, evolutionary strategies, simulated annealing, tabu search and simplex search) is performed. Further, a novel simulation annealing-based heuristic approach is proposed for critical infrastructure.
Findings
A small scale network of 50–100 nodes shows that genetic simulation optimization with multi-criteria and multi-dimensional features performs better as compared to other simulation optimization approaches. Further, a minimum of 3.4 percent and maximum of 16.2 percent improvement is observed in faster route identification for small scale Internet-of-things (IoT) networks with simulation optimization constraints integrated model as compared to the traditional method.
Originality/value
In this work, simulation optimization techniques are applied for identifying optimized Quality of service (QoS) parameters for critical infrastructure which in turn helps in improving the network performance. In order to identify optimized parameters, Tabu search and ant-inspired heuristic optimization techniques are applied over QoS parameters. These optimized values are compared with every monitoring sensor point in the network. This comparative analysis helps in identifying underperforming and outperforming monitoring points. Further, QoS of these points can be improved by identifying their local optimum values which in turn increases the performance of overall network. In continuation, a simulation model of bus transport is taken for analysis. Bus transport system is a critical infrastructure for Dehradun. In this work, feasibility of electric recharging units alongside roads under different traffic conditions is checked using simulation. The simulation study is performed over five bus routes in a small scale IoT network.
Details
Keywords
Jorge Luis Morato, Sonia Sanchez-Cuadrado, Christos Dimou, Divakar Yadav and Vicente Palacios
– This paper seeks to analyze and evaluate different types of semantic web retrieval systems, with respect to their ability to manage and retrieve semantic documents.
Abstract
Purpose
This paper seeks to analyze and evaluate different types of semantic web retrieval systems, with respect to their ability to manage and retrieve semantic documents.
Design/methodology/approach
The authors provide a brief overview of knowledge modeling and semantic retrieval systems in order to identify their major problems. They classify a set of characteristics to evaluate the management of semantic documents. For doing the same the authors select 12 retrieval systems classified according to these features. The evaluation methodology followed in this work is the one that has been used in the Desmet project for the evaluation of qualitative characteristics.
Findings
A review of the literature has shown deficiencies in the current state of the semantic web to cope with known problems. Additionally, the way semantic retrieval systems are implemented shows discrepancies in their implementation. The authors analyze the presence of a set of functionalities in different types of semantic retrieval systems and find a low degree of implementation of important specifications and in the criteria to evaluate them. The results of this evaluation indicate that, at the moment, the semantic web is characterized by a lack of usability that is derived by the problems related to the management of semantic documents.
Originality/value
This proposal shows a simple way to compare requirements of semantic retrieval systems based in DESMET methodology qualitatively. The functionalities chosen to test the methodology are based on the problems as well as relevant criteria discussed in the literature. This work provides functionalities to design semantic retrieval systems in different scenarios.
Details
Keywords
Balaji V., Kaliappan S., Madhuvanesan D.M., Ezhumalai D.S., Boopathi S., Patil Pravin P. and Saiprakash Mani
The purpose of the study is to examine the influence of the corn biofuel on the Jet engine. Each tests were carried out in a small gas turbine setup. The performance…
Abstract
Purpose
The purpose of the study is to examine the influence of the corn biofuel on the Jet engine. Each tests were carried out in a small gas turbine setup. The performance characteristics of thrust, thrust-specific fuel consumption, exhaust gas temperature and emission characteristics of Carbon monoxide(CO), Carbon dioxide (CO2), Oxygen (O2), Unburned hydrocarbons (UHC) and Nitrogen of oxides (NO) emissions were measured and compared with Jet-A fuel to find the suitability of the biofuel used.
Design/methodology/approach
Upgrading and using biofuels in aviation sector have been emerging as a fruitful method to diminish the CO emission into the atmosphere. This research paper explores the possibility of using nanoparticles-enriched bio-oil as a fuel for jet engines. The biofuel taken is corn oil and the added nanoparticles are Al2O3.
Findings
The biofuel blends used are B0 (100% Jet-A fuel), B10 (10 % corn oil biofuel + 90% Jet-A fuel), B20 (20% corn oil biofuel + 80% Jet-A fuel) and B30 (30% corn oil biofuel + 70% Jet-A fuel). All fuel blends were mixed with the moderate dosage level of 30 ppm. All tests were conducted at different rpm as 50,000, 60,000, 70,000 and 80,000 rpm.
Originality/value
The results proved that within the lower limit, use of biofuel increased the performance characteristics and reduced the emission characteristics except the emission of NO. The moderate-level biofuel with Jet-A fuel showed the equally better performance to the neat Jet-A fuel.