A Semantically-Enhanced Personalization Framework for ...

A Semantically-Enhanced Personalization Framework for ...

Personalized Information Retrieval in Context David Vallet Universidad Autnoma de Madrid, Escuela Politcnica Superior,Spain Overview Motivation Ontology-Based Content Retrieval Personalization Personalization in Context Building a Semantic Runtime Context Contextual Preference Activation Conclusions Motivation Requirements of two different multimedia applications european research projects: digital album (aceMedia) and a news service (MESH) Indicate users preferences

Content High level: Topics Low level: Topic sub-categories Geographical area Ontology-Based Preference Representation Personalised content Search results Browsing Context awareness Temporal preference Different scopes Session focused interests Personalisation in Context Ontology-Based Content Retrieval Info

need Formal query Goal: Improve keyword-based search Query engines Inference engines Ontology KB Annotation Documents Search space Returned document s Ranking ? Ontology-Based Content Retrieval Query q similarity d , q cos Documents

x2 d1 d2 x1 q x2 d1 x3 Ontology q d1 d2 x1 q1 d11 d21 x2 q2

d12 d22 x3 q3 d13 d23 1 2 d2 x1 x3 {x1, x2, x3} = domain ontology Personalization Users Preferences/Context Ontology KB Annotation

Documents Search space Personalization Personalization effect x2 score d , u , q f sim d , q , sim d , u f score d , q ,cos d1 u 1 2 d2 x1

x3 {x1, x2, x3} = domain ontology Personalization Ontology-Based Preference Representation Concepts VS Keywords Interoperability Precision Hierarchical Representation Inference Personalization

Ontology-Based Preference Representation C C Topics C Politics C Sports C Leisure visit C Travel C Island Travel C Movies C Music C Techno C Classical C Pop C Region Geographical Region C Islands C Spanish Islands C USA Islands I Hawaii Hawaii Tourist Guide Political Region C America

C NorthAmerica C Canada C USA locatedIn C Florida C Personalisation in Context Combination of long-term (preferences) + short-term (context) user interests and needs Not all user preferences are relevant all the time: which ones? Partial answer: focus on current semantic context, discard out of context ones Notion of context Defined as the set of background themes under which user activities occur within a given unit of time Represented as a set of weighted ontology concepts involved in user actions within a session

Captured? Build a runtime context: extracting concepts from queries and documents selected by the user Used? Contextual preference activation: Analyze semantic connections between preference and context concepts Personalization retrieval in context: Filter user preferences, only those related to the context are activated Building a Runtime Context Concepts, t Context Contextt t Contextt Action Action

Query Query t Content viewed Content annotations Content modified Action Action Query Query Query Visual query Textual query Query concepts Visual feedback Concept average

11 concepts Contextual Preference Activation preference for x = px r (x,y) preference for y = px w (r) Constrained Spreading Activation px 0.8 pyy w (r) 0.5 0.4 = 0.8 + 0.5 0.724 = 0.4 (1 - 0.4) 0.9 0.6 C Beach x

C nextTo r Sea y 0.9 nee ds 0.6 C Boat Domain ontology Domain ontology Contextual Preference Activation Domain concepts Semantic user preferences Initial user preferences Extended user preferences Contextt Contextualised

user preferences Initial runtime context Extended context Personalization in Context score d , u , q f sim d , q , sim d , u f score d , q ,cos score d , u, q, t f sim d , q , sim d , u , t f sim d , q , cos ' x2 d1

1 1 C uu, t 22 d 2 x1 x3 {x1, x2, x3} = domain ontology

Conclusions Semantic concepts VS plain terms Exploitation of semantic relation Semantic runtime context Context: Filtering of user preference References Semantic Search Personalization D. Vallet, M. Fernndez, P. Castells, P. Mylonas, and Y. Avrithis. Personalized Information Retrieval in Context. 3rd International Workshop on Modeling and Retrieval of Context (MRC 2006) at the 21st National Conference on Artificial Intelligence (AAAI 2006). Boston, USA, July 2006. Ranking Aggregation D. Vallet, P. Mylonas, M. A. Corella, J. M. Fuentes, P. Castells, and Y. Avrithis. A SemanticallyEnhanced Personalization Framework for Knowledge-Driven Media Services. IADIS

WWW/Internet Conference (ICWI 2005). Lisbon, Portugal, October 2005. Personalization in context P. Castells, M. Fernndez, and D. Vallet. An Adaptation of the Vector-Space Model for Ontology-Based Information Retrieval. IEEE Transactions on Knowledge and Data Engineering, 2007. In press. M. Fernndez, D. Vallet, and P. Castells. Using Historical Data to Enhance Rank Aggregation. 29th Annual International ACM Conference on Research and Development on Information Retrieval (SIGIR 2006), Poster Session. Seattle, WA, August 2006. Tuning Personalization P. Castells, M. Fernndez, D. Vallet, P. Mylonas, and Y. Avrithis. Self-Tuning Personalized Information Retrieval in an Ontology-Based Framework. 1st IFIP WG 2.12 & WG 12.4 International Workshop on Web Semantics (SWWS 2005), November 2005. Springer Verlag Lecture Notes in Computer Science, Vol. 3762. Meersman, R.; Tari, Z.; Herrero, P. (Eds.), 2005, ISBN: 3-540-29739-1, pp. 977-986. Thank You!

Recently Viewed Presentations

  • PLASMA MEMBRANE Submitted by Dr Madhurima Sharma LIPID

    PLASMA MEMBRANE Submitted by Dr Madhurima Sharma LIPID

    PLASMA MEMBRANE Submitted by Dr Madhurima Sharma
  • Quantitative Chemical Analysis 7e

    Quantitative Chemical Analysis 7e

    In general, analytes with equal concentrations give different detector responses. Therefore, the response must be measured for known concentrations of each analyte. A graph of detector response as a function of analyte concentration is called a .
  • CALVIN: Optimization of Californias Water Supplies 29 November

    CALVIN: Optimization of Californias Water Supplies 29 November

    PowerPoint Presentation Water Demand Functions Calvin's Economic Requirements Data Base for the SWAP Model PowerPoint Presentation SWAP Model Regions PowerPoint Presentation Agricultural Crop Descriptions Agricultural Response to Changes in the Price and Quantity of Water PowerPoint Presentation Efficiency-Cost Trade-offs.
  • Braden Score + Skin Inspection + Interventions +

    Braden Score + Skin Inspection + Interventions +

    Braden Score + Skin Inspection + Interventions + Critical Thinking Pressure Ulcer Prevention Objectives: Identify risks of pressure ulcer development Differentiate between skin assessment and skin inspection Discuss interventions for pressure ulcer prevention Test your accuracy at scoring a Braden...
  • Zing: Concurrent Software MC - cs.ioc.ee

    Zing: Concurrent Software MC - cs.ioc.ee

    Arial Times New Roman Symbol Tahoma Comic Sans MS Wingdings 2 SimSun Courier New Default Design ZING Systematic State Space Exploration of Concurrent Software Lecture III: Outline Stuck-Free Conformance Motivation Behavioral Contracts Goal Contract Checking Contract Checking Contracts Slide 10...
  • Airfoils, Lift and Bernoulli's Principle

    Airfoils, Lift and Bernoulli's Principle

    However, there can be too much of a good thing. The airfoil's ability to create lift is dependent on the airflow remaining smooth. Think of a stream flowing gently around a rock. The water's flow changes direction to go around/over...
  • Example of a scientific poster - Marquette University

    Example of a scientific poster - Marquette University

    Frumin E, Schlang J, Wiechmann W, et al. Prospective Analysis of Single Operator Sonographic Optic Nerve Sheath Diameter Measurement for Diagnosis of Elevated Intracranial Pressure. ...
  • Titration Excellence - Microsep

    Titration Excellence - Microsep

    Overview. Titration is an analytical method which is widely used in many different segments and for many different samples. Chemical industry: Water content in solvents, acid value in resins and polymers, hydroxyl value in polyols, surfactant content in raw materials,...