Detecting Favorite Topics in Computing Scientific Literature via Dynamic Topic Modeling

Topic modeling comprises a set of machine learning algorithms that allow topics to be extracted from collection documents. These have been widely used in many areas, such as the identification dominant scientific research. However, works addressing problems often focus on identifying static topics, providing snapshots are unable show how those evolve over time. Aiming close this gap, article, we describe an approach for dynamic article analysis and classification. This is accomplished by querying open data notable databases via representational state transfers subsequently enforcing management practices with topic associated metadata available. As result, identify research trends given field at specific instants referred terminology evolution throughout years. It was possible detect lexical variation time published content, ultimately determining so-called “hot topics” arbitrary also they correlated each other.