Open AccessA Robust Cybersecurity Topic Classification Tool
Author(s) -
Elijah Pelofske,
Lorie M. Liebrock,
Vincent Urias
Publication year - 2022
Publication title -
international journal of network security and its applications/international journal of network security and applications
Language(s) - English
Resource type - Journals
eISSN - 0975-2307
pISSN - 0974-9330
DOI - 10.5121/ijnsa.2022.14101
Subject(s) - computer science , scalability , the internet , artificial intelligence , machine learning , false positive rate , task (project management) , order (exchange) , mechanism (biology) , computer security , world wide web , database , engineering , systems engineering , finance , economics , philosophy , epistemology
In this research, we use user defined labels from three internet text sources (Reddit, StackExchange, Arxiv) to train 21 different machine learning models for the topic classification task of detecting cybersecurity discussions in natural English text. We analyze the false positive and false negative rates of each of the 21 model’s in cross validation experiments. Then we present a Cybersecurity Topic Classification (CTC) tool, which takes the majority vote of the 21 trained machine learning models as the decision mechanism for detecting cybersecurity related text. We also show that the majority vote mechanism of the CTC tool provides lower false negative and false positive rates on average than any of the 21 individual models. We show that the CTC tool is scalable to the hundreds of thousands of documents with a wall clock time on the order of hours.