Internet news hot event automatic generation system and method

Abstract

The invention relates to an internet news hot event automatic generation system and method, and the system comprises a data reading unit, a text word segmentation unit, a stop word removal unit, a keyword extraction and similarity calculation unit, an event forming and event abstract and title generation unit, an event classification unit and an event early warning unit. Wherein the data reading unit is respectively connected with the text word segmentation unit, the stop word removal unit, the keyword extraction unit, the similarity calculation unit and the event formation unit and transmits event data information, and the event formation unit is connected with the title generation unit, the event classification unit and the event early warning unit and transmits event information. According to the invention, the problem of automatic generation of Internet news hotspots is solved on the whole; compared with the prior art, the method has the advantages that the accuracy of automatic event generation is improved by adopting a new process; and the large-batch text processing rate is improved.

Description

technical field [0001] The invention relates to a technology in the field of information processing, in particular to an automatic generation system and method of Internet news hot events. Background technique [0002] The era of the Internet has spawned the development of many new media. People's sources of information are no longer as single as before. Except for new media, everyone can express their opinions on the Internet. If the new media is the source of information, then the comments of the masses are an important factor affecting the direction of events. Moreover, in the era of big data, news topics spread at an unimaginable speed. When a piece of news breaks out, there may be tens of thousands of retweets and millions of readings in just a few minutes. Such massive amounts of information can be disseminated explosively. How to grasp the public sentiment in real time and make corresponding processing is crucial for many enterprises and even government agencies. ...

AI Technical Summary

Problems solved by technology

If the extraction of keywords is not accurate enough, the formed summary will be biased, and it will not be able to present the beginning and end of the event well, which will eventually lead to poor accuracy of the entire system and no competition

[0007] The second is that existing technologies mostly use traditional processing methods in the process of event clustering

News is a typical streaming text, and the clustering of news events is time-sensitive, so these clustering algorithms are not suitable for automatic generation of Internet news hot events

[0009] The third is that the titles generated by existing technologies are not readable and cannot accurately express the original text

[0010] The title is the most important part of an event. It is the user's initial impression and understanding of the event. If the title cannot summarize the full text well, or even the extracted title is less readable, it will cause poor user experience. Consequences The traditional extractive headline generation just makes the headlines with low readability and poor generalization ability

Images