GeniUS: Generic User Modeling Library for the Social Semantic Web

DelftUniversity ofTechnology

GeniUS: Generic User Modeling Library for the Social Semantic WebJIST2011, December 2011, Hangzhou, China

Qi Gao, Fabian Abel, Geert-Jan Houben {q.gao, f.abel, g.j.p.m.houben}@tudelft.nl

Web Information SystemsDelft University of Technology

2GeniUS: Generic User Modeling Library for the Social Semantic Web

PersonalizedRecommendations

Personalized Search Adaptive Systems

What we do: Science and Engineering for the Personal Web

Social Web

Analysis and User Modeling

user/usage data

Semantic Enrichment, Linkage and Alignment

domains: news social media cultural heritage public data e-learning

3GeniUS: Generic User Modeling Library for the Social Semantic Web

Motivation• Sparsity problem

• do not have enough useful information for a (new) user

• Possible solution: gathering user data from other sources

User Modeling

Product Recommender

I’m a new user. Recommend me

some product

?• But not all data may be relevant for the given application context.• how to filter out user data that does not fit the target application context?

?

4GeniUS: Generic User Modeling Library for the Social Semantic Web

Research Challenges of GeniUS

Analysis and User Modeling

Semantic Enrichment

Productrecommender

Profile

?

Various applications in different domains

Movie recommender

Hotel recommender

interested in:

MovieProduct location

customized user profile construction

Movielocation

Productrecommender

How can we build a flexible and extensible user modeling functionality that adapts to the demands of a given

application context?

5GeniUS: Generic User Modeling Library for the Social Semantic Web

What is GeniUS?

• GeniUS is a topic and user modeling software library that• produces semantically meaningful profiles to enhance the

interoperability of profiles between applications;• provides functionality for aggregating relevant information

about a user from the Social Web;• generates domain-specific user profiles according to the

information needs of different applications;

• is flexible and extensible to serve different applications.

6GeniUS: Generic User Modeling Library for the Social Semantic Web

GeniUS: Generic Topic and User Modeling Library for the Social Semantic Web

Item Fetcher

Enrichment

Weighting

Function

RDF Repository

Filter

ModelingConfiguratio

n

RDF Serializati

on

Social Web

Semantic Web

user data itemsenriched

items

semantic data

user profiles

interested in:

locationproduct

7GeniUS: Generic User Modeling Library for the Social Semantic Web

GeniUS modules: Item Fetcher and Semantic Enrichment

Item Fetcher

Enrichment

Social Web

Twitter API

raw content a <sioc:Post> ; dcterms:created … ; sioc:has_creator …; sioc:content … .

Awesome, love the new Garageband for iPad #apple

Awesome, love the new Garageband for iPad #apple

SpotLight,Zemanta,

OpenCalais

sioc:has_topic dbpedia:Apple_Inc;sioc:has_topic dbpedia:GarageBand;sioc:has_topic dbpedia:Ipad;

dbpedia:GarageBanddbpedia:Ipaddbpedia:Apple_Inc

Garageband iPad #apple

8GeniUS: Generic User Modeling Library for the Social Semantic Web

Weighting Function

RDF Serializati

on

TFTF-IDF

Time-sensitiveRDF

Serialization

weight(dbpedia:GarageBand)

weight(dbpedia:Second_Life)

weight(dbpedia:Jazz)

the weighted interests vocabulary

GeniUS modules: Weighting Function and RDF Serialization

9GeniUS: Generic User Modeling Library for the Social Semantic Web

(Jazz, 0.5889)

(Second_Life, 0.3114)

(GarageBand, 0.1638)

GeniUS modules: Configuration and Filter

FilterModeling

Configuration

ModelingConfiguratio

n

itemsenriched

items

Filter

Twitter

APISpotLig

htTF

(Second_Life, 0.4101)

(GarageBand, 0.2158)

SELECT DISTINCT ?t WHERE { ? <rdf:type> <dbpedia-owl:Software> }

10GeniUS: Generic User Modeling Library for the Social Semantic Web

GeniUS: Generic Topic and User Modeling Library for the Social Semantic Web

Social Web

Semantic Web

GeniUS User Profileinterested in:

location …product

Applications

How do user profiles generated by GeniUS supportdifferent types of applications?

11GeniUS: Generic User Modeling Library for the Social Semantic Web

Analysis of Domain-specific User Profile Construction

• Dataset• 72 Twitter users (CS researchers) observed over a period of 6

months (> 40,000 tweets)• a variety of topics mentioned in the tweets

• Research questions• 1. What are the characteristics of (complete) Twitter-based

profiles generated with GeniUS ?

• 2. Can domain-specific profiles be derived from Twitter activities ?

• 3. What are the characteristics of such domain-specific profiles?

12GeniUS: Generic User Modeling Library for the Social Semantic Web

Analysis of Domain-specific User Profile Construction

average number of entities: 1097.1

a potential to generate domain-specific profilesby categorizing entities according to their types

average number of types: 35.0

13GeniUS: Generic User Modeling Library for the Social Semantic Web

Analysis of Domain-specific User Profile Construction

Are the domain-specific user profiles beneficial for supporting different recommendation applications?

×generic (all domains)

domain: locationdomain: entertainment domain: product

the more specific the domain the smaller the profiles

14GeniUS: Generic User Modeling Library for the Social Semantic Web

Evaluation of Domain-specific User Profile Construction

• Task: Recommending domain-specific tweets• Domains:

• three domains: location, entertainment, product• three sub-domains of product: book, software, music

• Recommender algorithm: cosine similarity between profile and candidate item

• Ground truth: relevant (re-)tweets of users • Candidate items: all the tweets posted during

evaluation period

time

P(u)= ?

1 month

Recommendations = ?

user profile

15GeniUS: Generic User Modeling Library for the Social Semantic Web

Evaluation results

the domain-specific user modeling strategies improve the performance of recommendations

three different domains

16GeniUS: Generic User Modeling Library for the Social Semantic Web

Evaluation results

The user modeling quality varies only slightly between the different domains

three sub-domainsof product

The sub-domain-specific user modeling strategy also improve the performance of recommendation.

17GeniUS: Generic User Modeling Library for the Social Semantic Web

Wrap up• GeniUS: Generic topic and User modeling library for the Social

Semantic Web• exploits traces (e.g. tweets) that people leave on the Social Web• enriches the semantics of these traces • constructs semantic user profiles profile construction can be customized and is adapted to a given application

context

• Analysis: • Twitter-based user profiles contain a great variety of topics• GeniUS succeeds in generating profiles for different applications and

domains

• Evaluation:• domain-specific user modeling strategies (powered by the semantic

filtering of GeniUS) allow clearly for the best performance• the more GeniUS adapts to the given domain (and application context) the

better the performance

18GeniUS: Generic User Modeling Library for the Social Semantic Web

Qi Gaoq.gao@tudelft.nlTwitter: @qigaoshhttp://wis.ewi.tudelft.nl/tweetum/http://wis.ewi.tudelft.nl/genius/

Thank You!