Almost 80% of all data is generated in the real world, which generates a large volume of unstructured data. This unstructured data require complex processing methods, in contrast to organised data stored in databases. Text mining is an effective method for knowledge extraction and decision-making since it includes transforming unstructured text input into structured format. Classification, clustering, association rule mining, topic detection, and summarising are the main text mining tasks. This book examines cutting-edge text mining techniques and uses, such as customer relationship management, social network analysis, financial market forecasting, and document classification. Document classification, which incorporates programmes like phishing, malware, and spam detection, is one of the difficult challenges in text classification.

The book suggests brand-new hybrid models for document categorization that work for both binary class and one-class problems, such as OCSVM-LSI, PCA-OCSVM, and LDA-CARM. The difficult task of stock market forecasting, which depends on unstructured news data, is also covered in the book. The suggested model predicts stock prices using a variety of regression approaches, including GMDH, GRNN, MLP, RPART, SVR, RF, and QRRF. The book provides insightful information on text mining's uses and how to make good decisions with it.