Analysis represented in vector space model is often used in information retrieval, topic analysis, and automatic classification. However, it hardly deals with fuzzy information and decision-making problems. To account this, Intuitionistic partition based cosine similarity measure between topic/terms and correlation between document/topic are proposed for evaluation. Conceptual granulation is emphasized in the decision matrix expressed conventionally as tf-idf. A local clustering of topic-terms and document-topics results in comparing dependent terms with membership degree using cosine similarity measure and correlation. A preprocessing of documents with intuitionistic fuzzy sets results in efficient classification of large corpus. But it depends on the datasets chosen. The proposed method effectively works well with large sized categorized corpus.
Document Analysis represented in vector space model is often used in information retrieval, topic analysis, and automatic classification. However, it hardly deals with fuzzy information and decision-making problems. To account this, Intuitionistic partition based cosine similarity measure between topic/terms and correlation between document/topic are proposed for evaluation. Conceptual granulation is emphasized in the decision matrix expressed conventionally as tf-idf. A local clustering of topic-terms and document-topics results in comparing dependent terms with membership degree using cosine similarity measure and correlation. A preprocessing of documents with intuitionistic fuzzy sets results in efficient classification of large corpus. But it depends on the datasets chosen. The proposed method effectively works well with large sized categorized corpus.