OLAP Query Logs for Proactive Personalization Julien Aligon 1 - PowerPoint PPT Presentation

ADBIS’2011 Mining Preferences from OLAP Query Logs for Proactive Personalization Julien Aligon 1 – Matteo Golfarelli 2 – Patrick Marcel 1 – Stefano Rizzi 2 – Elisa Turricchia 2 1 Université François Rabelais Tours Laboratoire Informatique France 2 University of Bologna DEIS Italy Session 3.A – September 26 th 2011

2 Mining Preferences from OLAP Query Logs for Proactive Personalization Motivation MDX query ADBIS’2011

3 Mining Preferences from OLAP Query Logs for Proactive Personalization Motivation MDX query ADBIS’2011 MDX Query myMDX [TKDE 2011] PREFERRING [ICDE 2011] Prescriptiveness Formulation Effort Proactiveness Expressiveness    

4 Mining Preferences from OLAP Query Logs for Proactive Personalization Motivation MDX query ADBIS’2011 MDX Query myMDX [TKDE 2011] PREFERRING Profile inferred from the context [ICDE 2011] and/or past actions. Prescriptiveness Formulation Effort Proactiveness Expressiveness    

5 Mining Preferences from OLAP Query Logs for Proactive Personalization Motivation MDX query ADBIS’2011 MDX Query myMDX [TKDE 2011] PREFERRING [ICDE 2011] Facts are ordered according to preferences Prescriptiveness Formulation Effort Proactiveness Expressiveness    

6 Mining Preferences from OLAP Query Logs for Proactive Personalization Motivation MDX query ADBIS’2011 MDX Query myMDX [TKDE 2011] PREFERRING Anticipate the user's [ICDE 2011] preference query Prescriptiveness Formulation Effort Proactiveness Expressiveness    

7 Mining Preferences from OLAP Query Logs for Proactive Personalization Motivation MDX query ADBIS’2011 MDX Query myMDX [TKDE 2011] PREFERRING Use of a rich language for [ICDE 2011] expressing preferences Prescriptiveness Formulation Effort Proactiveness Expressiveness    

8 Mining Preferences from OLAP Query Logs for Proactive Personalization Motivation MDX query ADBIS’2011 MDX Query myMDX PREFERRING Prescriptiveness Formulation Effort Proactiveness Expressiveness    

10 Mining Preferences from OLAP Query Logs for Proactive Personalization Proposition ADBIS’2011

11 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #1 How to model the query log? ADBIS’2011

12 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #1 How to model the query log?  A query is a set of fragments (Qf-set) ADBIS’2011 SELECT AvgIncome ON COLUMNS, MDX query: Crossjoin (OCCUPATION.members, Crossjoin ( Descendants (RACE.AllRaces,RACE.Mrn), Descendants (RESIDENCE.AllCities,RESIDENCE.Region))) ON ROWS FROM CENSUS WHERE TIME.Year.[2009] Measure: AvgIncome AllCities, Region Qf-set: AllRaces, Mrn Levels: Occ Year AllSex Selection: Year=2009  A log is a set of qf-set

13 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #2 What preferences can be extracted from the log? ADBIS’2011

14 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #2 What candidate preferences can be extracted from the log? What candidate preferences to extract ? ADBIS’2011  Rules of the form: context  candidate preference Qf-set (part of query) Single fragment How to extract these candidate preferences ?  Off-line extraction of association rules, using a classical algorithm (e.g., Apriori)  Confidence and support thresholds adjusted automatically, so that the set of extracted rules covers all the log

15 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #2 What candidate preferences can be extracted from the log? ADBIS’2011

16 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? ADBIS’2011

17 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? How candidate preferences of the log are found relevant? ADBIS’2011  By matching the rules of the log with the fragments of the user’s query q Not every rule is relevant for the user’s query:  Pertinent rule : the context is in the Qf-set of the query  Effective rule : the candidate preference is in the Qf-set of the query and allows to order the facts

18 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome ALLCITIES, REGION Qf-set: ALLRACES, MRN Levels: OCC ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 Answer to the query: (AllCities, AllRaces, Actors, 2009, AllSex, 15000) (Pacific, AllRaces, Actors, 2009, AllSex, 20000)

19 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome Non effective rule: ALLCITIES, REGION Qf-set: ALLRACES, MRN AllSex  Year=2009 Levels: OCC ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 Answer to the query: NO PREFERENCE (AllCities, AllRaces, Actors, 2009 , AllSex , 15000) (Pacific, AllRaces, Actors, 2009 , AllSex , 20000)

20 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome Pertinent and ALLCITIES, REGION Qf-set: effective Rule: ALLRACES, MRN Levels: OCC ADBIS’2011 YEAR Year=2009  Region ALLSEX Selection: YEAR=2009 Answer to the query: PREFERRED ( Pacific , AllRaces, Actors, 2009 , AllSex, 20000) (AllCities, AllRaces, Actors, 2009 , AllSex, 15000)

21 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome ALLCITIES, REGION Qf-set: ALLRACES, MRN Levels: OCC ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 Extracted rules of the log:

22 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome ALLCITIES, REGION Qf-set: ALLRACES, MRN Levels: OCC ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 1 st step : remove non pertinent (r 1 ) and non effective (r 5 , r 7 )

23 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome ALLCITIES, REGION Qf-set: nb of preferences to ALLRACES, MRN Levels: OCC add in the query : α =2 ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 • Region: 0.70 • AllCities: 0.60 2 nd step : group by • AvgIncome ∈ [500, 1000]: 0.55 candidate preference • Mrn: 0.45 • Year: 0.40

24 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome ALLCITIES, REGION Qf-set: ALLRACES, MRN Levels: OCC ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 • Region: 0.70 • AllCities: 0.60 3 rd step : select • AvgIncome ∈ [500, 1000]: 0.55 relevant fragment • Mrn: 0.45 • Year: 0.40

25 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome ALLCITIES, REGION Qf-set: ALLRACES, MRN Levels: OCC ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 • Region: 0.70 • AllCities: 0.60 3 rd step : select • AvgIncome ∈ [500, 1000]: 0.55 relevant fragment • Mrn: 0.45 • Year: 0.40

26 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome ALLCITIES, REGION Qf-set: ALLRACES, MRN Levels: OCC ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 • Region: 0.70 • AllCities: 0.60 3 rd step : select • AvgIncome ∈ [500, 1000]: 0.55 relevant fragment • Mrn: 0.45 • Year: 0.40

27 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #3 What preferences are relevant for the current query ? Measure: AvgIncome ALLCITIES, REGION Qf-set: ALLRACES, MRN Levels: OCC ADBIS’2011 YEAR ALLSEX Selection: YEAR=2009 • Region: 0.70 • AllCities: 0.60 3 rd step : select • AvgIncome ∈ [500, 1000]: 0.55 relevant fragment • Mrn: 0.45 • Year: 0.40

28 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #4 How to apply the relevant preferences to the query? ADBIS’2011

29 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #4 How to apply the relevant preferences to the query? How to translate the relevant candidate preference fragments into the query?  By using the preference constructor defined by myMDX ADBIS’2011

30 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #4 How to apply the relevant preferences to the query? ADBIS’2011 AvgIncome ∈ [500, 1000] • 4 th step : translate the BETWEEN(AvgIncome, 500, 1000) fragments • Mrn: 0.45 CONTAIN(Race, Mrn)

31 Mining Preferences from OLAP Query Logs for Proactive Personalization Issue #4 How to apply the relevant preferences to the query? How to combine preference constructors? ADBIS’2011  By AND clause between each successive constructors ( Pareto combination)  Each preference has the same importance