Research group

Machine Learning and Information Management Lab

Topic modeling of a news stream

25 April 2019

With the overwhelming growth of daily news Internet publications, there is an increasing need for automatic techniques to analyze news content.

This task can be stated as problem of topic modeling. During the seminar problem statement of topic modeling and brief review of methods for its solution will be given.

Further, we will discuss the restrictions of traditional approaches in short text modeling and clarify the task of news texts topic modeling.

In conclusion, the historical development of methods for analyzing the news texts structure and content will be disscussed as well as new approach will be presented.

Speaker: Ksenia Novikova.

Presentation language: Russian.

Date and time: April 25th, 8:00-9:30 pm.

Location: Times, room 405.

Videos from seminars is available at