The static, dynamic and incremental methods required an average of 7.7, 7.1 and 6.9 days, respectively, to detect and were able to precisely characterize the outbreak based on the detected topic (e.g., top keywords for ICD-9 code 569.3 were 'rectal'bleed'ed', and 'bleeding'bleeding
Since the terms are the basic elements of topics, the first step to detect hot topics is to extract key terms.
Its intuitive application is the word frequency which is intensively used to detect hot topics and trends in the society.
We use a foundational machine learning method known as non-negative matrix factorization (NMF) to detect crime topics, statistical collections of words reflecting latent structural relationships among crime events.
Section 2 discusses the related work; Section 3 presents the approach to detect hot topics; Section 4 introduces the details of two case studies; Section 5 discusses the findings and implications for supporting informal learning; and section 6 concludes this paper.
The major challenges in short text understanding are that short texts usually do not have the correct syntax that traditional POS-taggers or parsing methods can utilize and that they lack sufficient content to support statistical approaches to detect hidden topics.
However, this kind of unsupervised learning method can only detect the topics, not the entities.
In this work, we present a system called PoliTwi, which was designed to detect emerging political topics (Top Topics) in Twitter sooner than other standard information channels.
To combat this challenge, their approach focuses on detecting hidden topics for the tweets and then suggests the use of those general topics as hashtags using a latent dirichlet allocation (LDA) model to facilitate better search.
To detect sentiment-aware topics, we attempt to utilize Joint Sentiment/Topic models, these models are achieved with Latent Dirichlet Allocation (LDA) based models.
