Text Mining Approach for Prediction of Tumor Using Ontology Based Particle Swarm Optimization with Clustering Techniques

摘要

Text mining with Particle Swarm Optimization (PSO) Clustering Techniques to build a tumor prediction scheme. The proposed prediction scheme is based on Historical medical Reports associated with Tumor data. This research approach provides Effective Clustering by using Semantic Similarity that is calculated in Historical medical Reports Annotation Process. The Clustering Techniques group the reports into unsupervised cluster based on the features of the medical Reports. The Document Clustering is done through PSO. A PSO with ontology model of Clustering Knowledge Representation based on Historical medical report documents is presented and Compared to the traditional Support vector machine (SVM) approach. The SVM Methods to carry out the Integration of Medical ontology and the Text mining techniques is accomplished of mining the potential patterns and categorize clinical medical reports. Proposed ontology based frame work provides improved performance and better clustering compared to the traditional SVM Clustering.