A Similarity Measure for Classification of Text Extracted from Image using Machine Learning

Text processing plays a very important role in acquisition of data, data mining and web search. Estimating the comparability between texts is a significant undertaking in the text handling field. In this work, a normalized gauge of the similarity between the two texts in terms of features is proposed. Maxwell Boltzmann's is used to find similarities between 2 documents within the corpus document. A corpus document as a master plan, unique files as containers, attributes as sub- containers, and individual words as particles. The proposed measurement method is called the Maxwell - Boltzmann Similarity Measure (MBSM) [1]. It is found in the distribution of feature values and the total number of non-zero elements between texts. It is indicated that MBSM satisfies all document parallel rating structures. The MBSM is installed in the K nearest neighbors (KNN). The proposed scale will be extended to measure similarities between the 2 sets of texts.