Apr 2, 2020  publication descriptionINTERNATIONAL JOURNAL OF INNOVATIONS IN SCIENTIFIC ENGINEERING (IJISE) e-ISSN: 2454-6402/ p-ISSN:2454-812X in Vol 11, January-June, 2020

Nowadays, IoT is used in every possible area; this paper shows a detailed and conceives of the Internet of things and colossal information accessible on it. The article explains the effects and significance of these wordings in the present situation. Information Analytics is a developing field that uses IoT and enormous information. A large portion of the associations starts developing tremendous information accessible on IoT for their vital choices. However, associations are thinking that its valuable, likewise confronting security difficulties to ensure their meaningful data. Organizations that neglect to keep up sensible security, it implies for associations, is that if they neglect to verify the existence cycle of their huge information situations, at that point, they may confront administrative results, notwithstanding the noteworthy brand harm that information breaks can cause. To secure different methods like cryptography and encryption, and so on, are utilized. For secure correspondence in IoT-type frameworks as of now requires numerous degrees of the design of security algorithm, which disheartens clients from actualizing assurance and frequently urges usefulness to be organized over security. This paper features new security-upgrading strategies that depend on recognizable proof and confirmation of the things in the IoT condition. Worldwide Data is on the ascent. By 2020, we would have a complex of the information we produce each day. This information would be created through a full cluster of sensors we are ceaselessly joining in our lives. Improvement: So we need to actualize a standard way to deal with hazard the board, which accepts that the trust limit is as of now characterized. What is absent in the peril cantered, and techno-driven methodology is everything identified with the administration of the trust, i.e., the new capacities and forms, and the new arrangements and structures required to grow the hazard limit.

Feb 12, 2020  publication descriptionINTERNATIONAL JOURNAL OF INVENTIONS IN ENGINEERING & SCIENCE TECHNOLOGY (IJIEST) e-ISSN: 2454- 9584/ p-ISSN:2454-8111 in Vol 06, January-December, 2020

A wide variety of tragedies happen over the globe; the forecast of such catastrophe is imperative for early safeguard and clearing process. The expectation of a quake could be accomplished used antecedents or seismographic information; however, all such strategies can be completed distinctly by the area specialists (seismologist). Information mining strategies have been utilized in a wide assortment of utilizations and different areas. It permitted the forecast of execution and expected movement, which empowers the inference of wise choices. Forecasting earthquake history information can be accomplished by utilizing information mining ideas. In this paper, an expectation model is proposed for envisioning seismic tremors by applying clustering and affiliation rule mining on quake history information. At first, the data is gathered, and they are bunched, this grouped information is passed to the next stage where successive examples are acquired by applying affiliation rule mining, at long last by utilizing the case, the future quakes are anticipated by performing rule coordinating. This paper focuses on a prescient model using noteworthy quake information and mining strategies, which predicts the fore coming earthquake. This expectation model can be utilized to anticipate different seismic occasions, and they can be used for making a forecast in various fields by using a fitting dataset.

May 1, 2019  publication descriptionINTERNATIONAL JOURNAL OF INVENTIONS IN ELECTRONICS & ELECTRICAL ENGINEERING (IJIEEE) e- ISSN: 2454-9592/ p-ISSN:2454-8081 in Vol 5, January-December, 2019

Objective: This study presents IoT security features, IoT security layers’ challenges with the application layer and network layer. Methods/Findings: Major security concerns of IoT development are caused by the heterogeneity of interconnected entities, as well as the incompatibility of the development and communication protocols used. Authentication, integrity, and availability are compromised by attacks as man-in-the-middle, replay, and denial of service attacks. Looking into the future, blockchain and software-defined network technologies, are promising to deliver a more secure IoT operational environment, and could, therefore, reduce associated cyber-attack instances. Application: Furthermore, IoT security solution with current IoT security arrangements incorporate trust foundation, a move towards an integrated design and architecture is also discussed.

Jan 1, 2019  publication descriptionINTERNATIONAL JOURNAL OF ADVANCES IN ENGINEERING RESEARCH (IJAER) e-ISSN: 2231-5152/ p-ISSN:2454-1796 in Vol 17, Issue 1, January, 2019

Objective: Academic advising requires a lot of expertise, time, and responsibility. To assist the human advisors efficiently, the upcoming computerized advising system is a necessity. Methods/Statistical Analysis: Course Advisory System has been implemented using the EKA tool to recommend subjects for 8th class students of the ICSE board. Machine learning algorithms – Naïve Bayes, J48, PART, Random Forest, and KNN, have been modeled and tested on the data set. The performance of each classifier has been compared and analyzed. Findings: It is inferred that no advising system has been developed to assist school students in subject selection. Research work based on Indian students’ requirements is minimal. Research work based on students’ data caters more to binary class problems, whereas the addressing of multiclass issues is minimal. This work proposes an advising system for the school students of the 8th standard of ICSE board to choose their electives. Application/Improvements: This work focuses on the Indian educational system of school students. The approach takes care of the school students, which will add its advantage to the existing systems. As school students are more vulnerable by making the wrong decisions, the course Advisory system will assist them in analyzing their academic history and help them choose their electives wisely. The classification algorithms might give better accuracy with increasing instances. The Course advisory system can be enhanced using an ensemble approach.

Dec 1, 2018  publication descriptionINTERNATIONAL JOURNAL OF RESEARCH IN SCIENCE & TECHNOLOGY (IJRST) International Referred Journal e-ISSN: 2249-0604/ p-ISSN:2454-180X in Vol 08, Issue 4, Oct-Dec, 2018

Objectives: To review various tools available for simulating Spiking Neural Networks using heterogeneous parallel processing platforms that help to reduce cost, increase the computational speed, and also to document/archive lessons learned. Methods/Statistical Analysis: The computational speed is a continuing challenge for simulating good spiking neural network models. Understanding of the spiking neural networks is significantly simplified by computer simulators like NEST, GeNN, EDLUT, and BRIAN. Findings: Simulation is a handy toolkit of scientists and engineers of all disciplines. NEST, GeNN, EDLUT, and BRIAN simulators help in achieving better performance not in terms of the same kind of processing but with additional particular tasks that require more computational power. BRIAN and EDLUT, which are hybrid simulators, supports both time-driven and event-driven techniques and outperform when compared to other simulators. Application/Improvements: Using BRIAN and EDLUT simulation techniques, we can achieve high performance when compared to other spiking neural simulation techniques

Feb 1, 2018  publication descriptionINTERNATIONAL JOURNAL OF INNOVATIONS IN APPLIED SCIENCES AND ENGINEERING (IJIASE) e-ISSN: 2454-9258/ p-ISSN:2454-809X in Vol 04, January-December, 2018

Objectives: To provide related literature on the detection of Abusive language on Twitter using natural language processing (NLP). Methods: In this study, the survey has been conducted on different techniques and research done on the types of Abusive language used in social media, why it is essential? How it has been detected in real-time social media platforms and the performance metrics that are used by researchers in evaluating the performance of the detection of abusive language on Twitter by the users. Results: Giving an organized review of past methodologies, including methods, essential features, and core algorithms, this study arranges and depicts the present condition about this area. The study also talks about the intricacy of hate speech ideas, which is characterized by numerous stages of ad settings. This area of research has obvious potential for societal effects, especially in digital media and online networks. A crucial step in propelling automatic hate speech detection is the advancement and systematization of shared assets, for example, clarified data sets in numerous dialects, rules, and calculations. Conclusion: This survey study contains all the relevant references related to the detection of abusive language on social media using NLP and machine learning methods. Ultimately, it can be a source of reference to the other researchers in finding the pieces of literature that are relevant to their research area in the detection of Abusive language on Twitter.