• Balaji C. Narwade
• Akash D.Surwase
• Sahil S.Zele
• Bhalchandra A. Patil

The text grouping consists of automatically organizing data instances into groups. The information should inform a user about the content of each group without further observation of the specific instances, allowing a user to quickly scan relevant data groups. We modelling the descriptive cluster as an auto-coder network that predicts the characteristics of cluster assignments and predicts the cluster assignments of a subset of characteristics. The subset of functionality used to predict a cluster serves as a domain analysis. For the text documents, word count, phrases or other parameters provides a representation of the dispersed features with interpretable feature labels. In the proposed system, cluster predictions are performed using logistic regression models and feature forecasts are based on Logistic regression models. The optimization of these models leads to a completely self-regulating descriptive grouping approach that automatically selects the number of clusters and the number of functions for each cluster. We apply the methodology to a variety of short PDF documents and has shown that the selected grouping, as demonstrated by the subsets of the selected features.

Document clustering, feature selection, model selection, machine learning

"Self-Tuned Descriptive Document Clustering using a Predictive Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.470-473, May-2019, Available :http://www.jetir.org/papers/JETIR1905E69.pdf

"Self-Tuned Descriptive Document Clustering using a Predictive Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp470-473, May-2019, Available at : http://www.jetir.org/papers/JETIR1905E69.pdf