CYBERBULLYING DETECTION ON SOCIAL MEDIA PLATFORMS USING MACHINE LEARNING
Abstract
The development of self-healing concrete has the potential to revolutionize the construction industry by significantly reducing maintenance costs and increasing the lifespan of concrete structures. Carbon nanotubes (CNTs) have emerged as promising materials for self-healing concrete due to their unique mechanical and electrical properties. This major project aims to develop a novel self-healing concrete using CNTs by synthesizing and characterizing CNTs, optimizing their dispersion within the concrete, and evaluating the resulting material's mechanical and self-healing properties. The project's methodology involves first synthesizing and characterizing CNTs through various techniques such as scanning electron microscopy (SEM), transmission electron microscopy (TEM), and X-ray diffraction (XRD). Then, the dispersion of CNTs in concrete will be optimized through ultrasonic and sonication methods. Finally, the mechanical and self-healing properties of the resulting material will be evaluated using various tests such as compression and flexural strength tests, as well as water permeability tests to assess the material's self-healing ability. The successful development of self-healing concrete using CNTs could lead to a significant reduction in maintenance costs and the increased lifespan of concrete structures, making them more sustainable and resilient. Furthermore, using CNTs in concrete could pave the way for the development of other high- performance construction materials. Overall, the project's results could have significant implications for the construction industry and contribute to the development of more sustainable and resilient infrastructure. Cyberbullying is a type of crime in which anonymous user will send, post and share negative content about something irrelevant. Cyberbullying is a serious issue that is being faced on the internet which is affecting teenagers and grown-ups. It has led to mis happenings like self-murder and depression. In the present world the social media platforms have become a most important part of one’s life. Communications on social media platforms has been increasing day by day. The given paper we use the data in two different forms of cyberbullying, hate speech tweets from Twitter and commentary grounded on personal attacks from Wikipedia forums to build a project that is used to detect cyberbullying in the text using Natural Language Processing and Machine learning. We are using various classifiers like SVM(Support vector machine),Random forest algorithm and Navies Bayes to provide more accuracy. For Wikipedia data the model provides preciseness above 80% and for Tweet data it gives preciseness above 90%.
Keywords
offensive language, Machine learning algorithms, Extracting features, Twitter, Wikipedia , Cyber bullying , Hate speech