similarity joins for strings and sets
DESCRIPTION
Similarity Joins for Strings and Sets. William Cohen. WHIRL approach:. SELECT R.a,S.a,S.b,T.b FROM R,S,T WHERE R.a=S.a and S.b=T.b. SELECT R.a,S.a,S.b,T.b FROM R,S,T WHERE R.a~S.a and S.b~T.b (~ TFIDF-similar). Query Q. WHIRL approach:. Link items as needed by Q. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/1.jpg)
Similarity Joins for Strings and SetsWilliam Cohen
![Page 2: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/2.jpg)
SELECT R.a,S.a,S.b,T.b FROM R,S,T
WHERE R.a=S.a and S.b=T.b
WHIRL approach:
![Page 3: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/3.jpg)
Link items asneeded by Q
WHIRL approach:
Query Q
SELECT R.a,S.a,S.b,T.b FROM R,S,T
WHERE R.a~S.a and S.b~T.b (~ TFIDF-similar)
R.a S.a S.b T.b
Anhai Anhai Doan Doan
Dan Dan Weld WeldIncrementally produce a ranked list of possible links, with “best matches” first. User (or downstream process) decides how much of the list to generate and examine.
William Will Cohen Cohn
Steve Steven Minton Mitton
William David Cohen Cohn
![Page 4: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/4.jpg)
WHIRL queries• Assume two relations:
review(movieTitle,reviewText): archive of reviews
listing(theatre, movieTitle, showTimes, …): now showing
The Hitchhiker’s Guide to the Galaxy, 2005
This is a faithful re-creation of the original radio series – not surprisingly, as Adams wrote the screenplay ….
Men in Black, 1997
Will Smith does an excellent job in this …
Space Balls, 1987
Only a die-hard Mel Brooks fan could claim to enjoy …
… …
Star Wars Episode III
The Senator Theater
1:00, 4:15, & 7:30pm.
Cinderella Man
The Rotunda Cinema
1:00, 4:30, & 7:30pm.
… … …
![Page 5: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/5.jpg)
WHIRL queries
• “Find reviews of sci-fi comedies [movie domain]
FROM review SELECT * WHERE r.text~’sci fi comedy’
(like standard ranked retrieval of “sci-fi comedy”)
• “ “Where is [that sci-fi comedy] playing?”FROM review as r, LISTING as s, SELECT *
WHERE r.title~s.title and r.text~’sci fi comedy’
(best answers: titles are similar to each other – e.g., “Hitchhiker’s Guide to the Galaxy” and “The Hitchhiker’s Guide to the Galaxy, 2005” and the review text is similar to “sci-fi comedy”)
![Page 6: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/6.jpg)
WHIRL queries• Similarity is based on TFIDF rare words are most important.• Search for high-ranking answers uses inverted indices….
The Hitchhiker’s Guide to the Galaxy, 2005
Men in Black, 1997
Space Balls, 1987
…
Star Wars Episode III
Hitchhiker’s Guide to the Galaxy
Cinderella Man
…
Years are common in the review archive, so have low weight
hitchhiker movie00137
the movie001,movie003,movie007,movie008, movie013,movie018,movie023,movie0031,…..
- It is easy to find the (few) items that match on “important” terms
- Search for strong matches can prune “unimportant terms”
![Page 7: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/7.jpg)
A* (best-first) search• Find shortest path between start n0 and goal ng:goal(ng)
– Define f(n) = g(n) + h(n)• g(n) = MinPathLength(n0 ,n)|• h(n) = estimate of path length from n to ng
– Algorithm:• OPEN= {n0 }• While OPEN is not empty:
– remove “best” (minimal f) node n from OPEN– if goal(n), output path n0n and stop– otherwise, add CHILDREN(n) to OPEN
» and record their MinPathLength parents
![Page 8: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/8.jpg)
empty circles = open set, filled = closed set; color = distance from the start (the greener, the further)
![Page 9: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/9.jpg)
A* (best-first) search• Find shortest path between start n0 and goal ng:goal(ng)
– Define f(n) = g(n) + h(n)• g(n) = MinPathLength(n0 ,n)|• h(n) = lower-bound of path length from n to ng
– Algorithm:• OPEN= {n0 }• While OPEN is not empty:
– remove “best” (minimal f) node n from OPEN– if goal(n), output path n0n and stop– otherwise, add CHILDREN(n) to OPEN
» and record their MinPathLength parents» …note this is easy for a tree
h is “admissible” and A* will always return the lowest-
cost path
![Page 10: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/10.jpg)
A* (best-first) search for best K paths• Find shortest path between start n0 and goal ng:goal(ng)
– Define f(n) = g(n) + h(n)• g(n) = MinPathLength(n0 ,n)|• h(n) = lower-bound of path length from n to ng
– Algorithm:• OPEN= {n0 }• While OPEN is not empty:
– remove “best” (minimal f) node n from OPEN– if goal(n), output path n0n
» and stop if you’ve output K answers– otherwise, add CHILDREN(n) to OPEN
» and record their MinPathLength parents» …note this is easy for a tree
h is “admissible” and A* will always return the K lowest-
cost paths
![Page 11: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/11.jpg)
Inference in WHIRL
• “Best-first” search: pick state s that is “best” according to f(s)
• Suppose graph is a tree, and for all s, s’, if s’ is reachable from s then f(s)>=f(s’). Then A* outputs the globally best goal state s* first, and then next best, ...
![Page 12: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/12.jpg)
Using A* For WHIRL Queries• Assume two relations:
review(movieTitle,reviewText): archive of reviews
listing(theatre, movieTitle, showTimes, …): now showing
The Hitchhiker’s Guide to the Galaxy, 2005
This is a faithful re-creation of the original radio series – not surprisingly, as Adams wrote the screenplay ….
Men in Black, 1997
Will Smith does an excellent job in this …
Space Balls, 1987
Only a die-hard Mel Brooks fan could claim to enjoy …
… …
Star Wars Episode III
The Senator Theater
1:00, 4:15, & 7:30pm.
Cinderella Man
The Rotunda Cinema
1:00, 4:30, & 7:30pm.
… … …
![Page 13: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/13.jpg)
A* search to solve WHIRL queries
• “Find reviews of sci-fi comedies [movie domain]FROM review SELECT * WHERE r.text~’sci fi
comedy’
• “ “Where is [that sci-fi comedy] playing?”FROM review as r, LISTING as s, SELECT * WHERE r.title~s.title and r.text~’sci fi
comedy’(best answers: titles are similar to each other –
e.g., “Hitchhiker’s Guide to the Galaxy” and “The Hitchhiker’s Guide to the Galaxy, 2005” and the review text is similar to “sci-fi comedy”)
review(Title,Text), Text ~ “sci fi comedy”
review(TitleA,Text),listing( TitleB,Where,When),Text ~ “sci fi comedy”,TitleA ~ TitleB
Answer to Q is an assignment Θ to all the variables in Q
![Page 14: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/14.jpg)
Using A* For WHIRL Queries
Menace II Society, 1993
In this urban crime drama, directed by twin brothers Allen and Albert Hughes, a young street hustler …
Men in Black, 1997
Will Smith does an excellent job in this campy sci-fi. With Jones playing straight man this comedy succeeds in …
Men in Tights, 1988
Only a die-hard Mel Brooks fan could claim to enjoy …
… …
The Kings Speech Impediment
The Senator Theater
1:00, 4:15, & 7:30pm.
Hunger Games
The Rotunda Cinema
1:00, 4:30, & 7:30pm.
Men In Black 3
Leows WaterFront
7:00 pm
review(TitleA,Text), listing(Where, TitleB,When),Text ~ “sci fi comedy”, TitleA ~ TitleB
![Page 15: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/15.jpg)
Using A* For WHIRL Queries
Menace II Society, 1993
In this urban crime drama, directed by twin brothers Allen and Albert Hughes, a young street hustler …
Men in Black, 1997
Will Smith does an excellent job in this campy sci-fi. With Jones playing straight man this comedy succeeds in …
Men in Tights, 1988
Only a die-hard Mel Brooks fan could claim to enjoy …
… …
The Black Stallion - Director’s Cut
The Senator
1:00, 4:15, & 7:30pm.
Hunger Games
The Rotunda Cinema
1:00, 4:30, & 7:30pm.
Men In Black 3
Leows WaterFront
7:00 pm
review(TitleA,Text), listing(TitleB,Where,When),Text ~ “sci fi comedy”, TitleA ~ TitleB
![Page 16: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/16.jpg)
Using A* For WHIRL Queries
Menace II Society, 1993
In this urban crime drama, directed by twin brothers Allen and Albert Hughes, a young street hustler …
Men in Black, 1997
Will Smith does an excellent job in this campy sci-fi. With Jones playing straight man this comedy succeeds in …
Men in Tights, 1988
Only a die-hard Mel Brooks fan could claim to enjoy …
… …
The Black Stallion - Director’s Cut
The Senator
1:00, 4:15, & 7:30pm.
Hunger Games
The Rotunda Cinema
1:00, 4:30, & 7:30pm.
Men In Black 3
Leows WaterFront
7:00 pm
review(TitleA,Text), listing(TitleB,Where,When),Text ~ “sci fi comedy”, TitleA ~ TitleB
![Page 17: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/17.jpg)
Using A* For WHIRL Queries
Menace II Society, 1993
In this urban crime drama, directed by twin brothers Allen and Albert Hughes, a young street hustler …
Men in Black, 1997
Will Smith does an excellent job in this campy sci-fi. With Jones playing straight man this comedy succeeds in …
Men in Tights, 1988
Only a die-hard Mel Brooks fan could claim to enjoy …
… …
The Black Stallion - Director’s Cut
The Senator
1:00, 4:15, & 7:30pm.
Hunger Games
The Rotunda Cinema
1:00, 4:30, & 7:30pm.
Men In Black 3
Leows WaterFront
7:00 pm
review(TitleA,Text), listing(TitleB,Where,When),Text ~ “sci fi comedy”, TitleA ~ TitleB
![Page 18: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/18.jpg)
A* (best-first) search for best K paths• Find shortest path between start n0 and goal ng:goal(ng)
– Define f(n) = g(n) + h(n)• g(n) = MinPathLength(n0 ,n)|• h(n) = lower-bound of path length from n to ng
– Algorithm:• OPEN= {n0 }, where n0 is an empty assignment to variables• While OPEN is not empty:
– remove “best” (minimal f) node n from OPEN– if goal(n), output path n0n
» and stop if you’ve output K answers– otherwise, add CHILDREN(n) to OPEN
» where CHILDREN(n) binds a few more variables
![Page 19: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/19.jpg)
A* (best-first) search for best K paths• Find shortest path between start n0 and goal ng:allVarsBound(ng)
– Define f(n) = g(n) + h(n)• g(n) = MinPathLength(n0 ,n)|• h(n) = lower-bound of path length from n to ng
– Algorithm:• OPEN= {n0 }, where n0 is an empty assignment to variables• While OPEN is not empty:
– remove “best” (minimal f) node n from OPEN– if allVarsBound(n), output path n0n
» and stop if you’ve output K answers– otherwise, add CHILDREN(n) to OPEN
» where CHILDREN(n) binds a few more variables
![Page 20: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/20.jpg)
A* (best-first) search for best K paths• Find shortest path between start n0 and goal ng:allVarsBound(ng)
– Define f(n) = g(n) + h(n)• g(n) = MinPathLength(n0 ,n)|• h(n) = lower-bound of path length from n to ng
– Algorithm:• OPEN= {n0 }, where n0 is an empty assignment θ and an empty “exclusion list” E• While OPEN is not empty:
– remove “best” (minimal f) node n from OPEN– if allVarsBound(n), output path n0n
» and stop if you’ve output K answers– otherwise, add CHILDREN(n) to OPEN
» where CHILDREN(n) binds a few more variables
![Page 21: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/21.jpg)
Inference in WHIRL
• Explode p(X1,X2,X3): find all DB tuples <p,a1,a2,a3> for p and bind Xi to ai.
• Constrain X~Y: if X is bound to a and Y is unbound, – find DB column C to
which Y should be bound– pick a term t in X, find
proper inverted index for t in C, and bind Y to something in that index
• Keep track of t’s used previously, and don’t allow Y to contain one.
![Page 22: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/22.jpg)
Inference in WHIRL
• Adding to exclusions E means that upper bound h decreases
• Looking at maxweight means that partial assignments that can’t match well are penalized
![Page 23: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/23.jpg)
![Page 24: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/24.jpg)
![Page 25: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/25.jpg)
![Page 26: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/26.jpg)
![Page 27: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/27.jpg)
![Page 28: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/28.jpg)
Aside: Why WHIRL’s query language was cool• Combination of cascading queries and getting top-k answers is very useful
– Highly selective queries: • system can apply lots of constraints• user can pick from a small set of well-constrained potential answers
– Very broad queries:• system can cherry-pick and get the easiest/most obvious answers• most of what the user sees is correct
• Similar to joint inference schemes• Can handle lots of problems
– classification,
![Page 29: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/29.jpg)
![Page 30: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/30.jpg)
![Page 31: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/31.jpg)
![Page 32: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/32.jpg)
![Page 33: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/33.jpg)
![Page 34: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/34.jpg)
![Page 35: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/35.jpg)
![Page 36: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/36.jpg)
![Page 37: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/37.jpg)
![Page 38: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/38.jpg)
Epilogue• A few followup query systems to WHIRL
–ELIXIR, iSPARQL, …• “Joint inference” trick mostly ignored
–and/or rediscovered over and over• Lots and lots of work on similarity/distance metrics and efficient similarity joins
–much of which rediscovers A*-like tricks
![Page 39: Similarity Joins for Strings and Sets](https://reader035.vdocument.in/reader035/viewer/2022062723/56813c20550346895da596d8/html5/thumbnails/39.jpg)
Outline• Why joins are important• Why similarity joins are important• Useful similarity metrics for sets and strings• Fast methods for K-NN and similarity joins
–Blocking– Indexing–Short-cut algorithms–Parallel implementation