The text mining has been identified as dominant throughout the era in many scientific problems. Number of techniques have been identified and proposed earlier, each uses different features towards mining text information. However, they suffer to achieve higher performance in mining relevant text information or documents. To support the development of text mining tasks, an efficient topical and informative measures based algorithm is presented, which uses semantic ontology and the taxonomy as dictionary. The text features of the documents has been extracted to generate set of terms. For any document, the text features are used to remove the noisy features like stop words, stemming and tagging. With the noise removed pure terms, the method estimates the Topical Depth Similarity (TDS) and informative Depth Similarity (IDM) measures. The measures has been estimated towards each document to perform text mining. The input query has been estimated for the TDS measure to identify the category of the query. According to the category of query, the method estimates the TDS and IDS measures to mining the text. The proposed method improve the performance of text mining with reduces false classification ratio.