• MITHUN B.RAMGIRWAR
• SHITAL Y. GAIKWAD

The descriptive grouping consists of automatically organizing data instances into groups. The description should inform a user about the content of each group without further examination of the specific instances, allowing a user to quickly scan relevant groups. Selection of descriptions is often based on heuristic criteria. 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 description. For the text documents, appearance or counting of words, phrases or other attributes provides a representation of the dispersed features with interpretable feature labels in the proposed network, cluster predictions are performed and feature forecasts are based on machine learning framework. 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 text documents and has shown that the selected grouping, as demonstrated by the subsets of the selected features, is associated with significant topical organization.

Self-tuned, Descriptive clustering, feature selection, machine learning.

"Survey Paper on Descriptive Document Clustering using on the basis of Predictive Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.7, Issue 6, page no.369-372, June-2020, Available :http://www.jetir.org/papers/JETIR2006394.pdf

"Survey Paper on Descriptive Document Clustering using on the basis of Predictive Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.7, Issue 6, page no. pp369-372, June-2020, Available at : http://www.jetir.org/papers/JETIR2006394.pdf