公开(公告)号：US20170364504A1

公开(公告)日：2017-12-21

申请号：US15183831

申请日：2016-06-16

Applicant: XEROX CORPORATION

Inventor： Sandipan Dandapat ,  Ragunathan Mariappan ,  Shourya Roy ,  Shreshtha Mundra

IPC: G06F17/27 ,  G06F17/24

CPC classification number: G06F17/279 ,  G06F17/241 ,  G06F17/2765 ,  H04M3/51 ,  H04M2203/401

Abstract: The disclosed embodiments illustrate methods data processing for real-time text analysis. The method includes receiving text content from a plurality of user-computing devices, wherein the text content comprises at least a current text segment and a previous text segment. The method further includes extracting one or more first features from the current text segment, wherein at least a first feature of the one or more first features corresponds to a difference between timestamps associated with each of the current text segment and the previous text segment. The method further includes categorizing the current text segment into a predetermined category of one or more predetermined categories, based on at least the one or more first features, automatically by utilizing a classifier. Further, the method includes predicting a likelihood of evolution of an attribute in the text content, based on the predetermined category associated with the current text segment.