Data analysis is a key procedure in the part of data science that obtains needed information from any statistics. The ease of access and maintenance makes structured data the most popular choice among many organizations even today. On the other hand, with the fast development of technology, large amounts of more unstructured data like text and image are being generated. Unstructured data is data that didn’t have any pre-defined system associated with it. Because of the accessibility of a high number of electronic text records from various sources explaining unstructured and semi-structured data, document categorization work becomes an interesting part to control data nature. Text classification is an efficient activity that can be achieved using the originality of categorization algorithms. Recently, Machine Learning (ML) approaches offer a novel chance to emerge unstructured data into existing knowledge bases without the requirement to manually organize the data into topic-based content enriched with semantic metadata. Hence in this work, Context aware extraction of concepts from unstructured data using Machine Learning algorithms is presented. Textured Context Pattern (TCP) method with Lexical Subgroup (LS) model is used to explain the relevancy between the feature of query document and from whole dataset. the Unsupervised Cross-Correlated Neural Network (CCNN) is used to find the matching feature. The performance of presented model is described regarding Precision, Recall and F1-score.

Structured data, Unstructured data, Machine Learning, Cross-Correlated Neural Network (CCNN).