Experimental results

Source code:
data set:

Table1.User social related data statistics

# sum average/user average/tweet
users count 6,907 —— ——
tweets numbers 10,160,317 1,471 ——
Include #Topic tweets 4,401,369 637 43%
Include @ tweets 6,301,995 912 62%
The number of words 175,040,499 25,342 17.2
Number of word types 541,430 78 ——
Number of concerns 10,810,638 1,565 ——

Table2.LDA, hLDA, Twitter-LDA and MS-LDA experimental results

 
LDA
hLDA
Twitter-LDA
MS-LDA
Sub-interest P R F P R F P R F P R F
Basketball 100.00 80.00 88.89 71.42 62.50 66.67 100.00 80.00 88.89 100.00 80.00 88.89
Baseball 100.00 100.00 100.00 55.55 50.00 52.63 100.00 100.00 100.00 100.00 100.00 100.00
Football 70.00 100.00 82.35 77.00 67.00 71.65 100.00 85.71 92.31 100.00 85.71 92.31
Volleyball 75.00 75.00 75.00 87.50 70.00 77.77 57.14 100.00 72.72 80.00 100.00 88.89
Physics 100.00 100.00 100.00 71.43 50.00 58.82 100.00 80.00 88.89 57.10 80.00 66.64
Biology 75.00 60.00 66.67 50.00 50.00 50.00 100.00 80.00 88.89 100.00 80.00 88.89
Astronomy 100.00 100.00 100.00 0.00 0.00 0.00 83.33 100.00 90.91 50.00 100.00 66.67
Psychology 0.00 0.00 0.00 75.00 75.00 75.00 83.33 100.00 90.91 100.00 100.00 100.00
Economy 33.30 66.70 44.42 71.00 63.00 66.67 33.33 66.67 44.44 50.00 100.00 66.67
Employment 100.00 80.00 88.89 100.00 100.00 100.00 50.00 80.00 61.54 100.00 100.00 100.00
Investing 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 66.67 50.00 57.14
Energy 67.70 100.00 80.02 100.00 100.00 100.00 0.00 0.00 0.00 80.00 100.00 88.89
Politics 100.00 60.00 75.00 50.00 46.00 47.92 100.00 60.00 75.00 100.00 60.00 75.00
History 50.00 50.00 50.00 0.00 0.00 0.00 100.00 75.00 85.71 60.00 75.00 66.67
Physics 100.00 100.00 100.00 71.43 50.00 58.82 100.00 80.00 88.89 57.10 80.00 66.64
Physics 100.00 100.00 100.00 71.43 50.00 58.82 100.00 80.00 88.89 57.10 80.00 66.64
Physics 100.00 100.00 100.00 71.43 50.00 58.82 100.00 80.00 88.89 57.10 80.00 66.64
Physics 100.00 100.00 100.00 71.43 50.00 58.82 100.00 80.00 88.89 57.10 80.00 66.64
Religion 55.60 71.47 50.00 50.00 50.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
Military 100.00 100.00 100.00 0.00 0.00 80.00 100.00 88.89 100.00 100.00 100.00 100.00
Movie 0.00 0.00 0.00 0.00 0.00 0.00 50.00 100.00 66.67 0.00 0.00 0.00
Music 60.00 100.00 75.00 100.00 100.00 100.00 75.00 100.00 85.71 50.00 100.00 66.67
Photography 80.00 100.00 88.89 88.23 75.00 81.08 83.00 100.00 90.71 80.00 100.00 88.89
Literature 0.00 0.00 0.00 100.00 100.00 100.00 50.00 60.00 54.55 100.00 100.00 100.00
Travel 83.00 100.00 90.71 50.00 50.00 50.00 100.00 80.00 88.89 100.00 80.00 88.89
Food 100.00 100.00 100.00 83.33 66.67 74.07 100.00 100.00 100.00 100.00 100.00 100.00
Climbing 100.00 100.00 100.00 100.00 100.00 100.00 57.10 100.00 72.69 100.00 100.00 100.00
Fishing 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
Average 68.69 73.82 69.89 61.69 53.90 59.27 75.09 81.14 76.18 82.24 87.11 82.96

Table3.MS-LDA extracts the high probability words for each interest

Interest Word
Interest 1 Travel, fish, climbing, trip, lake, boat, flight, vacation, …
Interest 2 Biology, scientific, physics, space, cell, stem, astronomy, Jupiter, …
Interest 3 Political, culture, civilization, populace, history, globalized, community, …
Interest 4 Theater, cultural, movie, music, entertainment, photography, multimedia, Oscar, …
Interest 5 Basketball, football, coach, athlete, Kobe, baseball, soccer, volleyball, …
Interest 6 Market, bank, oil, CEO, business, tax, economy, financial, stocks, …

Table4.MS-LDA extracts the high probability words for each child interest

Child Interest Word
Interest 1 NBA, Kobe, basketball, guards, Shaq , shoot, three-point, ESPN, …
Interest 2 trips, airlines, Disney, vacation, hotel, flight, travel, airport, …
Interest 3 Base, outfield, baseball, glove, bat, mask, pitcher, MLB, Babe Ruth, …
Interest 4 photo, view, shots, images, camera, art, capture, gallery, magazine, …
Interest 5 classical, jazz, songs, musician, music, tune, hip-hop, rock, Grammy, …
Interest 6 Market, bank, oil, CEO, business, tax, economy, financial, stocks, …