Text mining for proper identification of noun using Conceptual Dependency

Text mining is the way toward analyzing of vast accumulation of composed aspect to produce new data and to change unstructured content into organized information for use in further. The need of story summarization has raised its importance as stories seems very simple but incorporate many elementary things required in natural language processing. In the event to make system more insightful, while performing summarization of stories a major issue was proof of recognizable of noun because noun may refer high level of entities. Automated summarizations have many challenges like paraphrasing, understanding context based information impact, still its important due to its need and vast areas of application in email clients, report generation, news feed, Entity timeline, story line of events, sentence compression, event understanding etc. There is demand to develop such tools which capture prominent information from stories available on the web or offline. Study shows in summarization more frequently machine learning methods are used like naive bayes, SVM, hidden Markov model etc. In this paper more emphasis is given on the aspect of conceptual dependency on verb in stories, as verbs are words that shows action or the overall Physical condition of a person helping to take out the main drama behind the stories and also in identification in noun because it refers to particular entity like people or things. The conceptual dependency is mainly used to highlight the important event from stories based on verb and noun search for Enhancement of text mining.