Reusable AI-based ensemble model for detecting SQL injection in service-oriented architectures
Cybersecurity has become one of the most sought-after domains in the field of computer science. Protection of computing resources and information against disruptive cyber threats has garnered utmost attention in recent times, owing to the conventional methods used in the field that often fall short of detecting or preventing the ever-evolving collection of malware. With the advent of new technologies such as Machine learning and Artificial intelligence, it is possible to streamline the approaches in the field of Cybersecurity. These technologies can be used to detect and prevent malicious content, thereby developing successful security solutions. The right AI tech could help us process huge volumes of threat data, discover anomalies and effectively eliminate potential threats. Currently, the most common approach involves using regular expressions to sequentially compare the incoming request or its vector with a predefined set of signatures. Though this approach is widely prevalent, it falls short in terms of accuracy. This is due to the fact that the signatures are not updated often, and several logical problems or loops come up when regular expressions are used within thousands of individual rules. In this project, we aim to identify various injections among neutral input vectors using ML models and will be predicting whether the vectors are injections or not. An ensemble of a number of ML models is used to build a voting mechanism to have an accurate prediction. For the sake of demonstration, the application consists of a frontend built using react and a python flask backend server
Published by: Sudarshan M., Pranava B., Dr. G. S. Mamatha
Author: Sudarshan M.
Paper ID: V7I3-2088
Paper Status: published
Published: June 28, 2021
Number Script Recognition using Neural Networks
The ability for accurate digit recognizer modelling and prediction is critical for pattern recognition and security. A variety of classification machine learning algorithms are known to be effective for digit recognition. The purpose of this experiment is rapid assessment of multiple types of classification models on digit recognition problem. The work offers an environment for comparing four types of classification models in a unified experiment: Multi-class decision forest, Multi-class decision jungle, Multi-class Neural Network and Multi-class Logistic Regression. The work presents assessment results using 6 performance metrics: Overall accuracy, Average accuracy, Micro-averaged precision, Macro-averaged precision, Micro-averaged recall and Macro-averaged recall. The experimental results showed that the highest accuracy was obtained by a Multi-class Neural Network with a value of 97.14%. The purpose of this project was to introduce neural networks through a relatively easy-to-understand application to the general public. This paper describes several techniques used for preprocessing the handwritten digits, as well as a number of ways in which neural networks were used for the recognition task.
Published by: Y. Bhanu Prasad, A. Sai Kumar, Pruthvy Charan, Dr. G. Prasad Acharya
Author: Y. Bhanu Prasad
Paper ID: V7I3-2171
Paper Status: published
Published: June 28, 2021
AI Based license Plate recognition using CNN
The discovery based on the artificial intelligence of the Indian license plate is our theme. We have created a program that is able to take a photo from the surrounding area. At the end of the hardware, we need a pc (or raspberry pi) and a camera and at the end of the software, we need a library to download and process data (image). We have used OpenCV (4.1.0) and Python (3.6.7) for this project To get something (license plate) in the picture we need another tool that can see the Indian license plate to use Haar cascade, pre-trained on Indian license plates (to be updated soon) be YOLO v3). Our main objective is to establish a system that gives us the license plate number of a vehicle when given a low definition image captured by a surveillance camera at toll collection centres. Mostly our system is demanded for the purpose of traffic monitoring. Hence we designed a system that lessens the manual work of entering the license plate numbers. And also we built a system that increases the speed of processing toll collections or traffic violation punishments. Our model is built on convolutional neural networks where several mathematical computations are done within the six hidden layers and give the output characters using contour detection and character segmentation.
Published by: K V Yaswanth, Anvesh Donthi, A Venkata Ramana, Dr. M. Poornachandra Rao
Author: K V Yaswanth
Paper ID: V7I3-2167
Paper Status: published
Published: June 28, 2021
Age and Gender Detection using OpenCV
In this fast emerging world Artificial Intelligence plays a very vital role in every field of science . Everything is being automated from operating a remote to driving a car using Artificial Intelligence. We show a glimpse of such automated experience with this project. In this project we show how easy it is to detect faces and identify gender along with gender with the help of CNN(Convolutional Neural Networks) and OpenCV. Using these fields of Artificial Intelligence we can reduce the use of hardware components and complexities in this project. Along with CNN and OpenCV we use Adience dataset so that the output is achieved with accurate values in training and validation. For the output to be determined even with multiple parameters we use pre-trained model that is caffee model along with OpenCV. The proposed model can be used in surveillance purposes or in medical purposes.
Published by: Mahija Kante, Dr. Esther Sunandha Bandaru, Gadili Manasa, Meghana Emandi, Vanarasi Leela Lavanya
Author: Mahija Kante
Paper ID: V7I3-2163
Paper Status: published
Published: June 28, 2021
DNA sequencing using Nanomanufacturing
In this paper, we nanomanufacture for the first time a novel DNA sequencing using nanoscale hole, in synthetic single digit nanometer thickness membrane. The DNA sequencing is carried out for each base of DNA to sequence and detect by placing a multimeter and the readings are taken on the edges of the synthetic single digit nanometer thickness membrane. The multimeter readings gives the voltage change readings for each base of DNA as there will be a change in concentration in the presence of DNA inside a nanoscale hole, in synthetic single digit nanometer thickness membrane.
Published by: Vishal Nandigana, Sharad Kumar Yadav, D. Manikandan
Author: Vishal Nandigana
Paper ID: V7I3-2160
Paper Status: published
Published: June 28, 2021
AC nanopump design and manufacture
In this paper, we design and manufacture for the first time a novel AC nanopump to generate flow velocity. AC voltage is needed at nanoscale pump designs and manufacture because DC voltage nanopumps design and manufacture generate bubbles/cavitation/instability and are not scalable designs and are not scalable to manufacture. AC voltage driven nanopump designed in and manufactured demonstrated in this paper overcomes bubble generation/cavitation/instability and our AC nanopump is a scalable design and manufacture.
Published by: Vishal Nandigana, Sharad Kumar Yadav, Manikandan D., K. D. Jo, A. T. Timperman, N. R. Aluru
Author: Vishal Nandigana
Paper ID: V7I3-2146
Paper Status: published
Published: June 28, 2021
