Cyber bullying using machine learning
Loading...
Files
Date
2020-10-13T11:04:53Z
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Criminal minded casual discussions via web-based networking media (for example Twitter) shed light into their instructive encounters—assessments, emotions, and worries about the learning procedure. Information from such un-instrumented conditions can give important information to
illuminate understudy learning. Breaking down such information, notwithstanding, can be
testing. The unpredictability of criminal minded encounters reflected from internet based life content requires human translation. Not with standing, the developing size of information
requests programmed information examination procedures. In this information mining
calculation dependent on Naive Bayes Multi-Label Classifier is executed which contains a few stages like Data Collection from twitter, Cleaning the information by expelling stop words,
evacuation of non-letter and accentuation marks, likelihood of the words for different
classifications in particular Heavy Study Load, Sleep Problems, Lack of Social Engagement, Negative Emotion and Diversity Issues is evaluated. For all the tweets Accuracy, Precision,
Recall, F1 measure, Micro Averaged and Macro Averaged values are registered for every class
and further more for the different users. Consequently we can conclude on average what number of criminal disapproved have different classifications of issues just as stretch out this to the issues faced by the client.
Description
Keywords
1NH16CS014, 1NH16CS097, 1NH16CS082