ABSTRACT
Text analysis plays a significant role in information retrieval (IR) and prediction in real-time data processing platforms. There are certain issues in modeling an IR system such as document indexing, query evaluation, and system evaluation. In recent years, researchers have shown more interest toward the document indexing as the textual data has increased a lot such as social media, healthcare diagnosis, etc. To analyze and investigate the document indexing process, this chapter processes a set-based model with special reference to medical data. The main objective of this set-based model is to find the term weights for index terms for contextual medical term representation. The function of this model is to compute the similarity between a document and a query. Association rule-based evaluation provides the significance of the terms that exist in a document, which is adopted for classification. This seems to be time efficient and improves the mechanism of data retrieval and analysis. The set-based model improves the average precision of the answer 188set. In terms of computing performance, this model is also competitive and has more parametric executions. Also, this model can be easily understood and interpreted, and it suits real-time processing platforms. This chapter covers a set-based model for reinforcement learning design, a case study, the rank computation process, tools for evaluation, and their applications toward medical data analysis.
