29/11/2007Dutch-Belgian Database Day 2007 PAS: A Personal Alert System for Information Retrieval in CRISs Germán Hurtado Martín 1,2 Chris Cornelis 2 1. Hogeschool Gent, 2. Universiteit Gent
Dutch-Belgian Database Day /11/2007 Overview CRISs Fuzzy sets and Rough sets PAS project
Dutch-Belgian Database Day /11/2007 Overview CRISs Fuzzy sets and Rough sets PAS project
Dutch-Belgian Database Day /11/2007 CRISs: Current Research Information Systems Bring together information related to current research Publications, project descriptions, programmes, researchers, organizations, patents…
Dutch-Belgian Database Day /11/2007 Examples of CRISs USDA/CRIS: SICRIS: RIS: IWETO: Degóis: uniCRIS: euroCRIS:
Dutch-Belgian Database Day /11/2007 Information Retrieval in CRISs Fuzzy Rough
Dutch-Belgian Database Day /11/2007 Overview CRISs Fuzzy sets and Rough sets PAS project
Dutch-Belgian Database Day /11/2007 Fuzzy sets and rough sets Traditional approach: crisp sets Young people = {x People | 0<age(x)<27}
Dutch-Belgian Database Day /11/2007 Fuzzy sets and rough sets Fuzzy approach: fuzzy sets 0 if age(x) ≥ 30 1 if age(x) ≤ 20 (30 – age(x)) / 10 otherwise Young(x) =
Dutch-Belgian Database Day /11/2007 Fuzzy sets and rough sets Rough approach: rough sets Information system: (X, A) Equivalence relation in X: R Equivalence class of X: Rx Equivalence classes: {x1,x4}, {x2}, {x3}, {x5}, {x6} with P = {Organisat., Funding, Discipl.} {x1,x4,x5}, {x2}, {x3}, {x6} with P = {Organisation, Discipline} X A
Dutch-Belgian Database Day /11/2007 Rough set: representation X Upper approx. R A ( Ry ∩ A ≠ Ø ) Lower approx. R A ( Ry A ) A positive examples Equivalence class of R
Dutch-Belgian Database Day /11/2007 Rough set (R↓A, R↑A) : example Equivalence class: {x1,x4}, {x2}, {x3}, {x5}, {x6} with P = {Org., Fund., Discipl.} R↑A R↓AR↓A A A = {x1, x2, x3} R↓A = {x2, x3} R↑A = {x1, x2, x3, x4}
Dutch-Belgian Database Day /11/2007 Fuzzy rough sets Fuzzy approach on rough sets Fuzzy set A Fuzzy relation R R (x,y) Upper approximation (R↑A)(y) = T (R(x,y),A(y)) Lower approximation (R↓A)(y) = I (R(x,y),A(y))
Dutch-Belgian Database Day /11/2007 Fuzzy rough sets: application Query expansion Allows more results by using R↑A RProgrammingHardwareC++JavaLaptopAlgorithm Programming Hardware C Java Laptop Algorithm Query: “Programming” - Expanded query: {(“Programming”,1.0), (“C++”,0.8), (“Java”,0.8), (“Algorithm”,0.6)}
Dutch-Belgian Database Day /11/2007 Overview CRISs Fuzzy sets and Rough sets PAS project
Dutch-Belgian Database Day /11/2007 PAS-project What is the PAS-project? Personal Alert System (HoGent) Goal: to get the researcher’s attention on funding possibilities that match his/her profile Information: about researchers, projects, funding possibilities (grants etc.) → matching/collaboration Automation and intelligence
Dutch-Belgian Database Day /11/2007 PAS – How does it work? -Name -Staff number -Department(s) -Group -Date of creation of the profile -Last update of the profile -Percentage research time -Skills description -Diplomas -Publications -IWETO-keywords -Free keywords Fill in IWETO- taxonomy Thesaurus 1 User
Dutch-Belgian Database Day /11/2007 PAS – How does it work? -Reference -Title -Content -Attachment(s) -Level -Duration -Institution -Deadline -Address -Contact person -IWETO-keywords -Free keywords IWETO- taxonomy Messages
Dutch-Belgian Database Day /11/2007 PAS – How does it work? The IWETO-classification has 641 research fields: 5 at the 1st level, 31 at the 2nd level, 605 at the 3rd level 1 2 3
Dutch-Belgian Database Day /11/2007 PAS – How does it work? By adding “free keywords” we can refine the classification
Dutch-Belgian Database Day /11/2007 PAS – How does it work? Query: A = {k3} Expanded query: R↑A = {(k1,0.8), (k3,1.0), …} M1 → R2
Dutch-Belgian Database Day /11/2007 PAS – How does it work?
Dutch-Belgian Database Day /11/2007 PAS – Current implementation Prototype developed as master’s thesis at the Hogeschool Gent Basic algorithm using weights and their products and basic fuzzy rough query expansion 1 Basic profiles and messages Manual processing of feedback Skeleton for the final system 1 P. Srinivasan, M. E. Ruiz, D. H. Kraft, J. Chen: Vocabulary mining for information retrieval: rough sets and fuzzy sets, Information Processing and Management, 37(1) (2001) 15-38
Dutch-Belgian Database Day /11/2007 PAS – Future work Richer representation of profiles and messages Automation of the feedback mechanism Dealing with imprecision and words from different thesauri Dealing with ambiguity and incomplete profiles Tracking research activities for collaboration Automatic extraction of information from text files
Dutch-Belgian Database Day /11/2007 Thank you