Clinical Data Classification Using an Ensemble Approach Based on CNN and Bag-of-Words Approach

Key phrases convey the important concepts in the text documents. They are useful for document categorization, clustering, indexing, search and summarization and to know about the semantic similarity with other documents. In this research article, a novel method is introduced for extracting key phrases from the chemical literature where the entities and terminologies are not just restricted to chemistry, but they also consist of words from different scientific domains. Six categories of phrases are defined by the chemistry experts such as chemistry, physics, medicine, drugs, cells and toxins. In this work, the extraction of the above six types of phrases is explained using pattern-based bootstrapping approach.