Text mining, otherwise called Intelligent Text Analysis is a significant exploration region. It is extremely challenging to zero in on the most suitable data because of the great dimensionality of information. Highlight Extraction is one of the significant techniques in information decrease to find the main features. Processing a huge measure of information put away in an unstructured structure is a challenging undertaking. Feature extraction is one of the huge pre-processing techniques in information mining that registers features esteem in documents. Thus, productive element extraction techniques term frequency-inverse document frequency (TF-IDF) techniques are regularly used in term weighting. This issue can't mean the accommodation or significance of certain features and diminishes the productivity of characterization. The record server executes stop word expulsion, labelling, and the examination of polysemous words in a pre-processing methodology to make a competitor corpus. Weight-based Core Corpus TF-IDF (W2CTF-IDF) is proposed to the competitor corpus to assess the significance of words in a bunch of documents. The words named of high significance by W2CTF-IDF are remembered for a bunch of keywords, and the transactions of each document are made. The technique is assessed W2CTF-IDF to weight the terms on financial data. The experiments show that W2CTF-IDF further develops the performance evaluation of component extraction as indicated by the maximum worth of the F1 measure.

Feature extraction, Text Mining, TF-IDF, Weight-based Core Corpus TF-IDF and Weighting;