Title :
Content-based topic discovery of high-impact model
Author :
Yang, Yun ; Wu, Yanan
Author_Institution :
Sch. of Electr. & Inf. Eng., Shaanxi Univ. of Sci. & Technol., Xi´´an, China
Abstract :
Because the traditional method of extracting hot topics exists some defects, therefore this article focuses on the content of theme, looking for these words having high-impact on theme and connecting with highly relevant words, accordingly we can extract high-impact theme in forum. The algorithm give a reasonable weight to each word. Combining the characteristic that reply to pasts continually in forum with symptom discovery algorithm, we can calculate the influence of word spreading the theme and extract the high frequency words and key words.
Keywords :
Internet; data mining; content based topic discovery; high impact theme extraction; hot topics extraction; symptom discovery algorithm; Data mining; Entropy; Frequency; Information filtering; Information filters; Internet; Joining processes; Lab-on-a-chip; Search engines; Correlation between words; high-frequency words; high-impact theme; high-key words;
Conference_Titel :
Computer Engineering and Technology (ICCET), 2010 2nd International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-6347-3
DOI :
10.1109/ICCET.2010.5486282
