agRec : An Enhanced Tag EnT EnTagRec Recommendation System for - - PowerPoint PPT Presentation

▶

Oct 25, 2023 380 likes •666 views

agRec : An Enhanced Tag EnT EnTagRec Recommendation System for Software Information Sites Shaowei Wang, David Lo, Bogdan Vasilescu, Alexander Serebrenik @b_vasilescu @aserebrenik / department of mathematics and computer science 18/02/15

SLIDE 1

EnT EnTagRec agRec: An Enhanced Tag Recommendation System for Software Information Sites

Shaowei Wang, David Lo, Bogdan Vasilescu, Alexander Serebrenik @b_vasilescu @aserebrenik

SLIDE 2 / department of mathematics and computer science Page 2 18/02/15

SLIDE 3 / department of mathematics and computer science Page 3 18/02/15

SLIDE 4 / department of mathematics and computer science Page 4 18/02/15

SLIDE 5 / department of mathematics and computer science Page 5 18/02/15

SLIDE 6 / department of mathematics and computer science Page 6 18/02/15

SLIDE 7 / department of mathematics and computer science Page 7 18/02/15

???

SLIDE 8 / department of mathematics and computer science Page 8 18/02/15

EnTagRec

SLIDE 9 / department of mathematics and computer science Page 9 18/02/15

EnTagRec TagCombine

r@5 0.805 0.595 p@5 0.346 0.221 r@5 0.815 0.568 p@5 0.358 0.251 r@5 0.88 0.675 p@5 0.369 0.278 r@5 0.64 0.639 p@5 0.382 0.381

Xia et al. MSR’13

SLIDE 10 / department of mathematics and computer science Page 10 18/02/15

EnTagRec TagCombine

r@5 0.805 0.595 p@5 0.346 0.221 r@5 0.815 0.568 p@5 0.358 0.251 r@5 0.88 0.675 p@5 0.369 0.278 r@5 0.64 0.639 p@5 0.382 0.381

SLIDE 11

EnT EnTagRec agRec: How have we done it?

SLIDE 12

EnT EnTagRec agRec: How have we done it?

L-LDA [Ramage et al. 2009] tokenization, identifier splitting, stop words, stemming

SLIDE 13

EnT EnTagRec agRec: How have we done it?

SLIDE 14 / department of mathematics and computer science Page 14 18/02/15

I have Java daemon which I want to pass shell commands. For example…

P( | )?

SLIDE 15 / department of mathematics and computer science Page 15 18/02/15

Java daemon want pass shell command exampl daemon load configur possibl

P( | )?

Actually (preprocessing…)

SLIDE 16 / department of mathematics and computer science Page 16 18/02/15

Java daemon want pass shell command exampl daemon load configur possibl

P( | )?

Tags = nouns (phrases)

SLIDE 17 / department of mathematics and computer science Page 17 18/02/15

Java

P( )

|

daemon

P( )

|

shell

P( )

|

…

Estimate from the training data Combine to get P for the entire text

SLIDE 18 / department of mathematics and computer science Page 18 18/02/15

Java daemon want pass shell command exampl daemon load configur possibl …

P( )

|

…

P( )

|

…

P( )

|

SLIDE 19 / department of mathematics and computer science Page 19 18/02/15

Supercalifragilisticexpialidocious

SLIDE 20 / department of mathematics and computer science Page 20 18/02/15

Supercalifragilisticexpialidocious

0.85

SLIDE 21

EnT EnTagRec agRec: How have we done it?

α*BIC + β*FIC Train α and β

SLIDE 22

EnT EnTagRec agRec

is better than BI

BIC and FIC FIC separately

/ department of mathematics and computer science Page 22 18/02/15

EnTagRec BIC FIC

r@5 0.805 0.565 0.593 p@5 0.346 0.232 0.258 r@5 0.815 0.505 0.637 p@5 0.358 0.212 0.282 r@5 0.88 0.523 0.713 p@5 0.369 0.212 0.298 r@5 0.64 0.391 0.545 p@5 0.382 0.230 0.322

SLIDE 23

EnT EnTagRec agRec

is better than BI

BIC and FIC FIC separately

is better than Ta

TagCom Combi bine

/ department of mathematics and computer science Page 23 18/02/15

SLIDE 24 / department of mathematics and computer science Page 24 18/02/15

So, what was all this about?

SLIDE 25 / department of mathematics and computer science Page 25 18/02/15

SLIDE 26 / department of mathematics and computer science Page 26 18/02/15

SLIDE 27 / department of mathematics and computer science Page 27 18/02/15