The document describes a multidimensional expertise recommender tool that aims to connect users with experts based on multiple criteria beyond just expertise. It discusses related work in expertise recommendation that focuses only on expertise. The proposed tool uses an architecture that profiles candidates on skills, qualities, proximity, and availability. It identifies initial candidates, assesses them against user requirements, aggregates the assessments using a fuzzy linguistic approach, and ranks the candidates. The document outlines improvements to the user interface and next steps to integrate the tool into an online platform and improve the prototype.
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
A multidimensional expertise recommender tool
1. A multidimensional expertise recommender tool
Germ´an S´anchez-Hern´andez, Jennifer Nguyen, N´uria Agell, Cecilio Angulo
June 24th, 2015
2. Introduction
State of the art
Arquitecture
User interface
Conclusions
Outline
1 Introduction
2 State of the art
3 Arquitecture
4 User interface
5 Conclusions
2 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
3. Introduction
State of the art
Arquitecture
User interface
Conclusions
Motivation and framework
Outline
1 Introduction
2 State of the art
3 Arquitecture
4 User interface
5 Conclusions
3 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
4. Introduction
State of the art
Arquitecture
User interface
Conclusions
Motivation and framework
Motivation and framework
www.projectcollage.eu
Creative learning is social, collaborative and peer based.
Expand and foster interaction among users with different
backgrounds, opinions and levels of expertise
→ improvement of creativity.
Find people with expertise in some areas
(“right level of expertise” vs. “right person”)
4 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
5. Introduction
State of the art
Arquitecture
User interface
Conclusions
Related work
Outline
1 Introduction
2 State of the art
3 Arquitecture
4 User interface
5 Conclusions
5 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
6. Introduction
State of the art
Arquitecture
User interface
Conclusions
Related work
Related work of ERS
MITRE’s Expert Finder [Mattox et al., 1999]: experts in
topics. Associations author-term in internal documents.
Search by keywords.
NASA POPS [Grove and Schain, 2008]: filtering experts by
organisation, project or competency. Use of RDF, multiple
databases, semantic web.
IBM SmaillBlue [Lin et al., 2008; 2009]: expertise in emails and
chat, mapping search strings to keywords.
INDURE FacFinder [Fang et al., 2008]: information from
faculty profiles and homepages, all internal documents
indexed. Proximity of terms. Considers order, source and rank
of source.
StrangersRS [Guy et al., 2011]: recommendation of people with
similar interests but unfamiliar.
6 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
7. Introduction
State of the art
Arquitecture
User interface
Conclusions
Related work
Lacks
Just expertise
Information treatment:
Filtering candidates
Parameters required (weights, thresholds)
→ Premature discards
7 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
8. Introduction
State of the art
Arquitecture
User interface
Conclusions
Related work
Lacks
Just expertise
Information treatment:
Filtering candidates
Parameters required (weights, thresholds)
→ Premature discards
7 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
9. Introduction
State of the art
Arquitecture
User interface
Conclusions
Related work
Lacks
Just expertise
Information treatment:
Filtering candidates
Parameters required (weights, thresholds)
→ Premature discards
7 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
10. Introduction
State of the art
Arquitecture
User interface
Conclusions
Related work
Lacks
Just expertise
Information treatment:
Filtering candidates
Parameters required (weights, thresholds)
→ Premature discards
7 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
11. Introduction
State of the art
Arquitecture
User interface
Conclusions
Related work
Lacks
Just expertise
Information treatment:
Filtering candidates
Parameters required (weights, thresholds)
→ Premature discards
7 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
12. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Outline
1 Introduction
2 State of the art
3 Arquitecture
4 User interface
5 Conclusions
8 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
13. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Arquitecture
9 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
14. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Arquitecture
Profiling
Profiling module
Candidates’ profiles
Access to Collage User
Profile Service
Offline massive update
Online selective update
10 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
15. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Profiles
Skill Profiling CER
Expertise Areas of knowledge
Four qualitative levels
(none, high, medium, low)
Qualities/Subskills Other knowledge Specific tools
Proximity Ease to contact
Physical distance
High or low
Availability Current availability
Manual update
Four levels
11 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
16. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Arquitecture
Interaction
Interaction module
Interaction with the user
Translates preferences to
requirements
Explicit
Implicit (user profile)
Controls for selecting final
experts
12 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
17. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Arquitecture
Identification
Identification module
Initial list of candidates
Access to profiles
Feasible candidates
13 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
18. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Arquitecture
Selection
Selection module
Assessing each candidate
Aggregating assessments
Ranking and selection of
candidates
14 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
19. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Selection of candidates
Assessment
One assessment per requirement. Fuzzy distance between the
requirement and the fulfillment.
Expertise: Ae(ps, l) = min(ps ,l)
l .
Subskill: Aq(ps) = ps.
Proximity: distance between user and candidate.
Ap(pd ) =
1 − pd −md
Md −md
if high proximity is required,
pd −md
Md −md
otherwise.
Availability: distance to (high) required availability
Aa(pa) = pa. (1)
(ps , pd and pa are related to the profile of the candidate; l is the required
level of expertise; md and Md are the minimum and maximum distances
between departments.)
15 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
20. Introduction
State of the art
Arquitecture
User interface
Conclusions
Arquitecture
CER
Selection of candidates
Aggregation and Ranking
Operator: OWA.
Weights: linguistic quantifier
wi = Q
i
n
− Q
i − 1
n
, i = 1, . . . , n.
“most of”: Q(r) = r
1
2 .
16 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
21. Introduction
State of the art
Arquitecture
User interface
Conclusions
Outline
1 Introduction
2 State of the art
3 Arquitecture
4 User interface
5 Conclusions
17 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
22. Introduction
State of the art
Arquitecture
User interface
Conclusions
Current UI
Input form
18 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
23. Introduction
State of the art
Arquitecture
User interface
Conclusions
New UI
Input form
19 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
24. Introduction
State of the art
Arquitecture
User interface
Conclusions
Current UI
Results
20 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
25. Introduction
State of the art
Arquitecture
User interface
Conclusions
New UI
Results
21 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
26. Introduction
State of the art
Arquitecture
User interface
Conclusions
Outline
1 Introduction
2 State of the art
3 Arquitecture
4 User interface
5 Conclusions
22 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
27. Introduction
State of the art
Arquitecture
User interface
Conclusions
Conclusions and Next Steps
Expertise recommender to find the right expert.
Right level of expertise vs. right person.
Additional information used: qualities, proximity, availability.
Integration into an existing platform (end users’ affinity space,
Moodle).
Information available in end users’ systems.
Improving prototype.
Team forming.
23 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender
28. Introduction
State of the art
Arquitecture
User interface
Conclusions
Conclusions and Next Steps
Expertise recommender to find the right expert.
Right level of expertise vs. right person.
Additional information used: qualities, proximity, availability.
Integration into an existing platform (end users’ affinity space,
Moodle).
Information available in end users’ systems.
Improving prototype.
Team forming.
23 / 24 Germ´an S´anchez-Hern´andez JARCA 2015: Expertise Recommender