A method and device for determining the time point of hot content in streaming media

Abstract

The embodiment of the invention discloses a method and a device for determining a content explosion point in streaming media. The method comprises the steps of: dividing the streaming media into multiple time periods according to a preset rule; for each of the time periods, carrying out word segmentation on bullet screens in the time period; making a statistics on a word frequency of each word segmentation, and determining bullet screen thematic words corresponding to the time period; determining an occurrence frequency of the bullet screen thematic words corresponding to the time period; according to the occurrence frequency of the bullet screen thematic words corresponding to each time period and a stored frequency mean value, identifying the time periods in which the occurrence frequency is greater than the frequency mean value; and in the identified time periods, according to the occurrence frequency of the corresponding bullet screen thematic words at each time point in the time periods, determining the time point corresponding to a maximum occurrence frequency value as the content explosion point. By adopting the method and the device disclosed by the embodiment of the invention, the efficiency and the accuracy of determining the content explosion point in the streaming media can be improved.

Description

technical field [0001] The present invention relates to the technical field of streaming media, in particular to a method and device for determining the time point of hot content in streaming media. Background technique [0002] The determination of the time point of hot content in the existing streaming media, such as the viewing point of the video, mainly relies on manual editing and determination. Due to the different sensitivity of editors to the content, there may be errors between the set point of view and the focus of most users. , cannot correctly reflect the topics that users are really interested in; and once the time point of a single streaming media hot content is set, it will generally not be updated continuously. When some time-sensitive hot topics change, the corresponding hot content cannot be effective displayed. [0003] It can be seen that in the prior art, this method of relying on manual editing to determine the time point of hot content in streaming me...

AI Technical Summary

Problems solved by technology

[0002] The determination of the time point of the hot content in the existing streaming media, such as the viewing point of the video, etc., mainly depends on manual editing and determination. Because editors have different sensitivities to the content, there may be errors between the set point of view and the focus of most users. , cannot correctly reflect the topics that users are really interested in; and once the time point of a single streaming media hot content is set, it will generally not be continuously updated. When some time-sensitive hot topics change, the corresponding hot content cannot be effective show up

[0003] It can be seen that in the prior art, this method of relying on manual editing to determine the time point of hot content in streaming media is inefficient, has large errors and is not updated in time

Images