Extracting Topic Changes through Word Co-occurrence Graphs from Japanese Subtitles of VOD Lecture

Kobayashi, Nobuyuki; Koyama, Noboru; Shiina, Hiromitsu; Kitagawa, Fumio

doi:10.1109/iiai-aai.2012.71

Search citation statements

Order By: Relevance

Paper Sections

Select...

) Co-occurrence Word-graph1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2014

2018

Publication Types

Select...

Other3

Relationship

Self Cite0

Independent3

Authors

Journals

Cited by 3 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The co-occurrence word-graph method treats words in each micro-b10gging as initial nodes, linking nearby co occurrence words with high levels of co-occurrence with edges [13].In this paper, word x and word ya re linked when C(x, y) is greater than or equal to j3 that is the lower limit of co-occurrence. Each connected sub graphs of the co-occurr ence word-graph is a topic.…”

Section: ) Co-occurrence Word-graphmentioning

confidence: 99%

Hot topics detected from micro-bloggings based on word co-occurrence model

Cao

Chen

Zhang

et al. 2014

2014 IEEE Computers, Communications and IT Applications Conference

View full text Add to dashboard Cite

Micro-blogging services are used by millions of people around the world to get information and express their opinions. Detecting hot topics from Chinese micro bloggings has vast importance to discovering rumors and guiding public opinion. In order to solve the problem of massive pieces of information on micro-bloggings platform and the feature of micro-bloggins content such as short text, in this paper a model is put foward to detect hot topics from Chinese micro-bloggings based on word co-occurrence model. The experimental results show the model can efficiently detect hot topics from Chinese micro-bloggings.

show abstract

Section: ) Co-occurrence Word-graphmentioning

confidence: 99%