SlideShare uma empresa Scribd logo
1 de 25
Context-Aware Points of Interest
Suggestion with Dynamic Weather
Data Management
Matthias Braunhofer, Mehdi Elahi, Francesco Ricci,
and Thomas Schievenin
Faculty of Computer Science
Free University of Bozen – Bolzano, Italy

ENTER 2014 Research Track

Slide Number 1
Agenda
• Recommender Systems and Context-Awareness
• Related Work
• Weather-Aware Recommendations
• Experimental Evaluation
• Conclusions and Future Work

ENTER 2014 Research Track

Slide Number 2
Recommender Systems (RSs)
• Goal: recommend new, relevant items to users based on
their feedback on a sample of items (training set)
– Explicit feedback (ratings) vs. implicit feedback (purchase /
browsing history)

• Two basic technical approaches:
– Collaborative filtering (CF)
– Content-based

ENTER 2014 Research Track

Slide Number 3
Context is Essential
• Main idea: users can experience items differently
depending on the current contextual situation (e.g.,
season, weather, temperature, mood)
• Example:

ENTER 2014 Research Track

Slide Number 4
Context-Aware
Recommender Systems (CARSs)
• CARSs extend RSs beyond users and items to the contexts
in which items are experienced by users
– Rating prediction function is: R: Users x Items x Context 
Ratings

1

3
2

3

4
5
3

1
5

2

5
2

4

5
?
5

4
?

4

2
3

3
4
3

3

ENTER 2014 Research Track

5
5

3
3

5

Slide Number 5
Using Weather in RSs
• Hypothesis: weather conditions at places of interest
(POIs), together with past ratings for POIs under several
distinct weather conditions, can be used to improve the
choice satisfaction and perceived recommendation quality
• Example:

?
5

1
?
ENTER 2014 Research Track

Slide Number 6
Agenda

ENTER 2014 Research Track

Slide Number 7
Context-Aware
Matrix Factorization (CAMF)
• CAMF (Baltrunas et al., 2011) extends matrix factorization
by incorporating baseline parameters for contextual
conditions-item pairs, which capture the deviation of
items‘ ratings due to context
k

ˆ
ruic1 ,...,ck = i + bu + ∑ bic j + pu qiT
j=1

• Main limitations:

ī: average rating of item i
bu: baseline for user u
bicj: baseline for contextual

condition cj and item i
pu: latent factor vector of user u
qi: latent factor vector of item i

– Fails to produce personalized recommendations for new users
– Uses the weather condition around the user rather than the
weather conditions at POIs
ENTER 2014 Research Track

Slide Number 9
Mobile CARSs (1/2)
• liveCities (Martin et al., 2011): supports tourists by
sending them push-based notifications when they enter a
certain area and their context matches pre-defined
conditions
• Main limitations:
-

Considers only the
temperature and not the
weather
Uses pre-defined
recommendations rather than
a predictive model
Is only a research prototype
ENTER 2014 Research Track

Slide Number 10
Mobile CARSs (2/2)
• VISIT (Meehan et al., 2013): hybrid mobile RS that uses
several contextual factors (i.e., location, time, weather,
social media sentiment) to support tourist‘s decisionmaking process
• Main limitation:
-

System is only a proposal and
not yet implemented

ENTER 2014 Research Track

Slide Number 11
Agenda

ENTER 2014 Research Track

Slide Number 12
STS – South Tyrol Suggests

ENTER 2014 Research Track

Slide Number 13
Weather-Aware
Recommendations
• Main idea: use CAMF (Baltrunas et al., 2011) as starting
point and incorporate additional user attributes (i.e.,
gender, birth date and personality trait information)
• Advantage: can produce personalized recommendations
for users without ratings
CAMF
k

ˆ
ruic1 ,...,ck = i + bu + ∑ bic j + qiT ( pu +
j=1

∑

ya )

a∈A(u )

A distinct factor vector ya corresponds to each
user attribute to describe a user through the
set of user-associated attributes

ENTER 2014 Research Track

Slide Number 14
Phase 1: Training
• The system learns the parameters offline (i.e., once every
5 minutes) by minimizing the regularized squared error on
the set of known ratings K (the training set):
Prediction error

minb*,q*, p*,y*

∑

2

(u,i,c1 ,...,ck )∈K
k

ˆ
(ruic1 ,...,ck − ruic1 ,...,ck ) +

λ (b + ∑ b + qi + pu +
2
u

j=1

2
ic j

2

2

∑

2

ya )

a∈A(u )

Regularization term to avoid overfitting

• Minimization performed by stochastic gradient descent
ENTER 2014 Research Track

Slide Number 15
Phase 2: Recommendation
• Retrieve the weather and temperature values for the 116
municipalities of South Tyrol by querying Mondometeo
– Note: This information can be cached

• For each POI in the database:
– Look up the POI‘s location
– Assign the weather and temperature values retrieved for the
closest municipality
– Compute a rating prediction, considering the weather and
temperature conditions along with other known contextual
conditions

• Recommend the top-20 POIs to the user
ENTER 2014 Research Track

Slide Number 16
Agenda

ENTER 2014 Research Track

Slide Number 17
Experimental Methodology
• Live user study where our proposed weather-aware
system (STS) was compared with a variant (STS-S) that has
the same graphical UI but does not use the weather
context when generating recommendations
• We have designed a specific user task and used a
questionnaire for assessing the perceived
recommendation quality and choice satisfaction
(Knijnenburg et al., 2012)
• 54 subjects that were randomly divided in two equal
groups assigned to STS and STS-S (27 each)
ENTER 2014 Research Track

Slide Number 18
User Task (1/2)
• Users were supposed to:
– have an afternoon off and to look for attractions / events in South
Tyrol
– consider the contextual conditions relevant for them and to
specify them in the system settings
– browse the attractions / events sections and check whether they
could find something interesting for them
– browse the system suggestions (recommendations), and select
and bookmark the one that they believe fits their needs and
wants
– fill up a survey (Knijnenburg et al., 2012), which measures
perceived recommendation quality and choice satisfaction
ENTER 2014 Research Track

Slide Number 19
User Task (2/2)
• After this initial interaction, users had the opportunity to
double check the weather conditions at the bookmarked
POI by accessing on a computer the Mondometeo website
– Will this knowledge influence the users choice?
– Hypothesis: STS has exploited this information so the user should
have already chosen items that are "compatible" with the
weather

• Users were then asked whether they wanted to change
their preferred POI and bookmark another one, i.e., if
they believed that, because of the weather conditions at
the bookmarked POI, their previous choice was not
anymore considered to be appropriate
ENTER 2014 Research Track

Slide Number 20
Results (1/2)
Number of unsatisfied users, i.e., those that changed their bookmarked
POI after having double checked the weather conditions at the POI

ENTER 2014 Research Track

Slide Number 21
Results (2/2)

Choice satisfaction

Perceived rec. quality

Statement

STS
avg.

STS-S avg.

pvalue

1. I liked the items suggested by the system

4.0

3.7

0.20

2. The suggested items fitted my preference.

3.4

3.4

0.56

3. The suggested items were well-chosen
4. The suggested items were relevant.

3.5
3.7

3.3
3.2

0.13
0.04

5. The system suggested too many bad items.

2.9

2.7

6. I didn’t like any of the suggested items.

3.8

3.3

0.14
<
0.001

7. The items I selected were “the best among the worst”.
8. I like the item I’ve chosen.

3.1
4.6

2.8
4.3

0.20
0.02

9. I was excited about my chosen item.

4.0

3.7

0.03

10. I enjoyed watching my chosen item.

3.7

3.9

0.79

11. The items I watched were a waste of my time.

3.5

3.5

0.42

12. The chosen item fitted my preference.

3.9

4.0

0.71

ENTER 2014 Research Track

Slide Number 22
Agenda

ENTER 2014 Research Track

Slide Number 23
Conclusions
• STS = mobile CARS that recommends POIs using a set of
contextual factors that include the current weather
conditions at the recommended POIs
• Novelty is the usage of up-to-date weather data into a
matrix factorization algorithm to generate personalized
context-aware recommendations
• Usage of weather data significantly improves the users‘
perceived recommendation quality and choice satisfaction

ENTER 2014 Research Track

Slide Number 24
Future Work
• Extended analysis
– To better understand potential performance
differences among the compared CARSs, which may be
due to different usage of the weather contextual
factors
– To test our proposed weather-aware CARS with a
larger number of users and a larger rating dataset

ENTER 2014 Research Track

Slide Number 25
Questions?
Thank you.

ENTER 2014 Research Track

Slide Number 26

Mais conteúdo relacionado

Mais procurados

Active Learning in Collaborative Filtering Recommender Systems : a Survey
Active Learning in Collaborative Filtering Recommender Systems : a SurveyActive Learning in Collaborative Filtering Recommender Systems : a Survey
Active Learning in Collaborative Filtering Recommender Systems : a SurveyUniversity of Bergen
 
User Personality and the New User Problem in a Context-Aware Point of Interes...
User Personality and the New User Problem in a Context-Aware Point of Interes...User Personality and the New User Problem in a Context-Aware Point of Interes...
User Personality and the New User Problem in a Context-Aware Point of Interes...University of Bergen
 
Empirical Evaluation of Active Learning in Recommender Systems
Empirical Evaluation of Active Learning in Recommender SystemsEmpirical Evaluation of Active Learning in Recommender Systems
Empirical Evaluation of Active Learning in Recommender SystemsUniversity of Bergen
 
Tutorial: Context In Recommender Systems
Tutorial: Context In Recommender SystemsTutorial: Context In Recommender Systems
Tutorial: Context In Recommender SystemsYONG ZHENG
 
[WI 2017] Context Suggestion: Empirical Evaluations vs User Studies
[WI 2017] Context Suggestion: Empirical Evaluations vs User Studies[WI 2017] Context Suggestion: Empirical Evaluations vs User Studies
[WI 2017] Context Suggestion: Empirical Evaluations vs User StudiesYONG ZHENG
 
Continuous Evaluation of Collaborative Recommender Systems in Data Stream Man...
Continuous Evaluation of Collaborative Recommender Systems in Data Stream Man...Continuous Evaluation of Collaborative Recommender Systems in Data Stream Man...
Continuous Evaluation of Collaborative Recommender Systems in Data Stream Man...Dr. Cornelius Ludmann
 
Tutorial: Context-awareness In Information Retrieval and Recommender Systems
Tutorial: Context-awareness In Information Retrieval and Recommender SystemsTutorial: Context-awareness In Information Retrieval and Recommender Systems
Tutorial: Context-awareness In Information Retrieval and Recommender SystemsYONG ZHENG
 
Context-Aware Recommender System Based on Boolean Matrix Factorisation
Context-Aware Recommender System Based on Boolean Matrix FactorisationContext-Aware Recommender System Based on Boolean Matrix Factorisation
Context-Aware Recommender System Based on Boolean Matrix FactorisationDmitrii Ignatov
 
[RIIT 2017] Identifying Grey Sheep Users By The Distribution of User Similari...
[RIIT 2017] Identifying Grey Sheep Users By The Distribution of User Similari...[RIIT 2017] Identifying Grey Sheep Users By The Distribution of User Similari...
[RIIT 2017] Identifying Grey Sheep Users By The Distribution of User Similari...YONG ZHENG
 
Retweet Prediction with Attention-based Deep Neural Network
Retweet Prediction with Attention-based Deep Neural NetworkRetweet Prediction with Attention-based Deep Neural Network
Retweet Prediction with Attention-based Deep Neural NetworkGUANGYUAN PIAO
 
[ADMA 2017] Identification of Grey Sheep Users By Histogram Intersection In R...
[ADMA 2017] Identification of Grey Sheep Users By Histogram Intersection In R...[ADMA 2017] Identification of Grey Sheep Users By Histogram Intersection In R...
[ADMA 2017] Identification of Grey Sheep Users By Histogram Intersection In R...YONG ZHENG
 
Collaborative Metric Learning (WWW'17)
Collaborative Metric Learning (WWW'17)Collaborative Metric Learning (WWW'17)
Collaborative Metric Learning (WWW'17)承剛 謝
 
Temporal Learning and Sequence Modeling for a Job Recommender System
Temporal Learning and Sequence Modeling for a Job Recommender SystemTemporal Learning and Sequence Modeling for a Job Recommender System
Temporal Learning and Sequence Modeling for a Job Recommender SystemAnoop Kumar
 
[WI 2017] Affective Prediction By Collaborative Chains In Movie Recommendation
[WI 2017] Affective Prediction By Collaborative Chains In Movie Recommendation[WI 2017] Affective Prediction By Collaborative Chains In Movie Recommendation
[WI 2017] Affective Prediction By Collaborative Chains In Movie RecommendationYONG ZHENG
 
A Combination of Simple Models by Forward Predictor Selection for Job Recomme...
A Combination of Simple Models by Forward Predictor Selection for Job Recomme...A Combination of Simple Models by Forward Predictor Selection for Job Recomme...
A Combination of Simple Models by Forward Predictor Selection for Job Recomme...David Zibriczky
 

Mais procurados (20)

Context-Aware Points of Interest Suggestion with Dynamic Weather Data Management
Context-Aware Points of Interest Suggestion with Dynamic Weather Data ManagementContext-Aware Points of Interest Suggestion with Dynamic Weather Data Management
Context-Aware Points of Interest Suggestion with Dynamic Weather Data Management
 
Active Learning in Collaborative Filtering Recommender Systems : a Survey
Active Learning in Collaborative Filtering Recommender Systems : a SurveyActive Learning in Collaborative Filtering Recommender Systems : a Survey
Active Learning in Collaborative Filtering Recommender Systems : a Survey
 
User Personality and the New User Problem in a Context-Aware Point of Interes...
User Personality and the New User Problem in a Context-Aware Point of Interes...User Personality and the New User Problem in a Context-Aware Point of Interes...
User Personality and the New User Problem in a Context-Aware Point of Interes...
 
Empirical Evaluation of Active Learning in Recommender Systems
Empirical Evaluation of Active Learning in Recommender SystemsEmpirical Evaluation of Active Learning in Recommender Systems
Empirical Evaluation of Active Learning in Recommender Systems
 
Tutorial: Context In Recommender Systems
Tutorial: Context In Recommender SystemsTutorial: Context In Recommender Systems
Tutorial: Context In Recommender Systems
 
[WI 2017] Context Suggestion: Empirical Evaluations vs User Studies
[WI 2017] Context Suggestion: Empirical Evaluations vs User Studies[WI 2017] Context Suggestion: Empirical Evaluations vs User Studies
[WI 2017] Context Suggestion: Empirical Evaluations vs User Studies
 
Continuous Evaluation of Collaborative Recommender Systems in Data Stream Man...
Continuous Evaluation of Collaborative Recommender Systems in Data Stream Man...Continuous Evaluation of Collaborative Recommender Systems in Data Stream Man...
Continuous Evaluation of Collaborative Recommender Systems in Data Stream Man...
 
Tutorial: Context-awareness In Information Retrieval and Recommender Systems
Tutorial: Context-awareness In Information Retrieval and Recommender SystemsTutorial: Context-awareness In Information Retrieval and Recommender Systems
Tutorial: Context-awareness In Information Retrieval and Recommender Systems
 
Context-Aware Recommender System Based on Boolean Matrix Factorisation
Context-Aware Recommender System Based on Boolean Matrix FactorisationContext-Aware Recommender System Based on Boolean Matrix Factorisation
Context-Aware Recommender System Based on Boolean Matrix Factorisation
 
[RIIT 2017] Identifying Grey Sheep Users By The Distribution of User Similari...
[RIIT 2017] Identifying Grey Sheep Users By The Distribution of User Similari...[RIIT 2017] Identifying Grey Sheep Users By The Distribution of User Similari...
[RIIT 2017] Identifying Grey Sheep Users By The Distribution of User Similari...
 
Analyzing User Reviews in Tourism with Topic Models
Analyzing User Reviews in Tourism with Topic ModelsAnalyzing User Reviews in Tourism with Topic Models
Analyzing User Reviews in Tourism with Topic Models
 
Retweet Prediction with Attention-based Deep Neural Network
Retweet Prediction with Attention-based Deep Neural NetworkRetweet Prediction with Attention-based Deep Neural Network
Retweet Prediction with Attention-based Deep Neural Network
 
[ADMA 2017] Identification of Grey Sheep Users By Histogram Intersection In R...
[ADMA 2017] Identification of Grey Sheep Users By Histogram Intersection In R...[ADMA 2017] Identification of Grey Sheep Users By Histogram Intersection In R...
[ADMA 2017] Identification of Grey Sheep Users By Histogram Intersection In R...
 
Contextual information elicitation in travel recommender systems
Contextual information elicitation in travel recommender systemsContextual information elicitation in travel recommender systems
Contextual information elicitation in travel recommender systems
 
Collaborative Metric Learning (WWW'17)
Collaborative Metric Learning (WWW'17)Collaborative Metric Learning (WWW'17)
Collaborative Metric Learning (WWW'17)
 
Recommender Systems
Recommender SystemsRecommender Systems
Recommender Systems
 
Temporal Learning and Sequence Modeling for a Job Recommender System
Temporal Learning and Sequence Modeling for a Job Recommender SystemTemporal Learning and Sequence Modeling for a Job Recommender System
Temporal Learning and Sequence Modeling for a Job Recommender System
 
[WI 2017] Affective Prediction By Collaborative Chains In Movie Recommendation
[WI 2017] Affective Prediction By Collaborative Chains In Movie Recommendation[WI 2017] Affective Prediction By Collaborative Chains In Movie Recommendation
[WI 2017] Affective Prediction By Collaborative Chains In Movie Recommendation
 
A Combination of Simple Models by Forward Predictor Selection for Job Recomme...
A Combination of Simple Models by Forward Predictor Selection for Job Recomme...A Combination of Simple Models by Forward Predictor Selection for Job Recomme...
A Combination of Simple Models by Forward Predictor Selection for Job Recomme...
 
Recommender Systems
Recommender SystemsRecommender Systems
Recommender Systems
 

Semelhante a Context-Aware Points of Interest Suggestion with Dynamic Weather Data Management

Geotecs: Exploiting Geographical, temporal, categorical, and social context f...
Geotecs: Exploiting Geographical, temporal, categorical, and social context f...Geotecs: Exploiting Geographical, temporal, categorical, and social context f...
Geotecs: Exploiting Geographical, temporal, categorical, and social context f...rameshraj
 
Recommending Sequences RecTour 2017
Recommending Sequences RecTour 2017Recommending Sequences RecTour 2017
Recommending Sequences RecTour 2017Gunjan Kumar
 
Usability evaluation methods (part 2) and performance metrics
Usability evaluation methods (part 2) and performance metricsUsability evaluation methods (part 2) and performance metrics
Usability evaluation methods (part 2) and performance metricsAndres Baravalle
 
Usability evaluations (part 2)
Usability evaluations (part 2) Usability evaluations (part 2)
Usability evaluations (part 2) Andres Baravalle
 
PERICLES presentation on Appraisal - IDCC15 workshop
PERICLES presentation on Appraisal - IDCC15 workshopPERICLES presentation on Appraisal - IDCC15 workshop
PERICLES presentation on Appraisal - IDCC15 workshopPERICLES_FP7
 
Context-aware Recommendation: A Quick View
Context-aware Recommendation: A Quick ViewContext-aware Recommendation: A Quick View
Context-aware Recommendation: A Quick ViewYONG ZHENG
 
Planning and usability evaluation methods
Planning and usability evaluation methodsPlanning and usability evaluation methods
Planning and usability evaluation methodsAndres Baravalle
 
Cognitive Science Perspective on User eXperience!
Cognitive Science Perspective on User eXperience!Cognitive Science Perspective on User eXperience!
Cognitive Science Perspective on User eXperience!Hamed Abdi
 
Advanced Methods for User Evaluation in Enterprise AR
Advanced Methods for User Evaluation in Enterprise ARAdvanced Methods for User Evaluation in Enterprise AR
Advanced Methods for User Evaluation in Enterprise ARMark Billinghurst
 
User Studies for APG: How to support system development with user feedback?
User Studies for APG: How to support system development with user feedback?User Studies for APG: How to support system development with user feedback?
User Studies for APG: How to support system development with user feedback?Joni Salminen
 
Thesis review Presentation
Thesis review PresentationThesis review Presentation
Thesis review PresentationAndrew Harvey
 
Gunjan insight student conference v2
Gunjan insight student conference v2Gunjan insight student conference v2
Gunjan insight student conference v2Gunjan Kumar
 

Semelhante a Context-Aware Points of Interest Suggestion with Dynamic Weather Data Management (20)

Geotecs: Exploiting Geographical, temporal, categorical, and social context f...
Geotecs: Exploiting Geographical, temporal, categorical, and social context f...Geotecs: Exploiting Geographical, temporal, categorical, and social context f...
Geotecs: Exploiting Geographical, temporal, categorical, and social context f...
 
Other metrics
Other metricsOther metrics
Other metrics
 
Recommending Sequences RecTour 2017
Recommending Sequences RecTour 2017Recommending Sequences RecTour 2017
Recommending Sequences RecTour 2017
 
Usability evaluation methods (part 2) and performance metrics
Usability evaluation methods (part 2) and performance metricsUsability evaluation methods (part 2) and performance metrics
Usability evaluation methods (part 2) and performance metrics
 
Usability evaluations (part 2)
Usability evaluations (part 2) Usability evaluations (part 2)
Usability evaluations (part 2)
 
PERICLES presentation on Appraisal - IDCC15 workshop
PERICLES presentation on Appraisal - IDCC15 workshopPERICLES presentation on Appraisal - IDCC15 workshop
PERICLES presentation on Appraisal - IDCC15 workshop
 
Information Retrieval Evaluation
Information Retrieval EvaluationInformation Retrieval Evaluation
Information Retrieval Evaluation
 
Context-aware Recommendation: A Quick View
Context-aware Recommendation: A Quick ViewContext-aware Recommendation: A Quick View
Context-aware Recommendation: A Quick View
 
User Personality and the New User Problem in a Context-­‐Aware POI Recommende...
User Personality and the New User Problem in a Context-­‐Aware POI Recommende...User Personality and the New User Problem in a Context-­‐Aware POI Recommende...
User Personality and the New User Problem in a Context-­‐Aware POI Recommende...
 
Planning and usability evaluation methods
Planning and usability evaluation methodsPlanning and usability evaluation methods
Planning and usability evaluation methods
 
201120 Yunha Han
201120 Yunha Han201120 Yunha Han
201120 Yunha Han
 
Cognitive Science Perspective on User eXperience!
Cognitive Science Perspective on User eXperience!Cognitive Science Perspective on User eXperience!
Cognitive Science Perspective on User eXperience!
 
Simplified planning technique
Simplified planning techniqueSimplified planning technique
Simplified planning technique
 
Towards a better understanding of the cognitive destination image of the Basq...
Towards a better understanding of the cognitive destination image of the Basq...Towards a better understanding of the cognitive destination image of the Basq...
Towards a better understanding of the cognitive destination image of the Basq...
 
Advanced Methods for User Evaluation in Enterprise AR
Advanced Methods for User Evaluation in Enterprise ARAdvanced Methods for User Evaluation in Enterprise AR
Advanced Methods for User Evaluation in Enterprise AR
 
User Studies for APG: How to support system development with user feedback?
User Studies for APG: How to support system development with user feedback?User Studies for APG: How to support system development with user feedback?
User Studies for APG: How to support system development with user feedback?
 
Why Do Travellers Change Their Trip?
Why Do Travellers Change Their Trip? Why Do Travellers Change Their Trip?
Why Do Travellers Change Their Trip?
 
Thesis review Presentation
Thesis review PresentationThesis review Presentation
Thesis review Presentation
 
Gunjan insight student conference v2
Gunjan insight student conference v2Gunjan insight student conference v2
Gunjan insight student conference v2
 
Temporal based Recommendation System
Temporal based Recommendation SystemTemporal based Recommendation System
Temporal based Recommendation System
 

Último

"Fly with Ease: Booking Your Flights with Air Europa"
"Fly with Ease: Booking Your Flights with Air Europa""Fly with Ease: Booking Your Flights with Air Europa"
"Fly with Ease: Booking Your Flights with Air Europa"flyn goo
 
Italia Lucca 1 Un tesoro nascosto tra le sue mura
Italia Lucca 1 Un tesoro nascosto tra le sue muraItalia Lucca 1 Un tesoro nascosto tra le sue mura
Italia Lucca 1 Un tesoro nascosto tra le sue murasandamichaela *
 
Apply Indian E-Visa Process Online (Evisa)
Apply Indian E-Visa Process Online (Evisa)Apply Indian E-Visa Process Online (Evisa)
Apply Indian E-Visa Process Online (Evisa)RanjeetKumar108130
 
Haitian culture and stuff and places and food and travel.pptx
Haitian culture and stuff and places and food and travel.pptxHaitian culture and stuff and places and food and travel.pptx
Haitian culture and stuff and places and food and travel.pptxhxhlixia
 
Authentic Travel Experience 2024 Greg DeShields.pptx
Authentic Travel Experience 2024 Greg DeShields.pptxAuthentic Travel Experience 2024 Greg DeShields.pptx
Authentic Travel Experience 2024 Greg DeShields.pptxGregory DeShields
 
Exploring Sicily Your Comprehensive Ebook Travel Guide
Exploring Sicily Your Comprehensive Ebook Travel GuideExploring Sicily Your Comprehensive Ebook Travel Guide
Exploring Sicily Your Comprehensive Ebook Travel GuideTime for Sicily
 
Inspirational Quotes About Italy and Food
Inspirational Quotes About Italy and FoodInspirational Quotes About Italy and Food
Inspirational Quotes About Italy and FoodKasia Chojecki
 
Aeromexico Airlines Flight Name Change Policy
Aeromexico Airlines Flight Name Change PolicyAeromexico Airlines Flight Name Change Policy
Aeromexico Airlines Flight Name Change PolicyFlyFairTravels
 
Revolutionalizing Travel: A VacAI Update
Revolutionalizing Travel: A VacAI UpdateRevolutionalizing Travel: A VacAI Update
Revolutionalizing Travel: A VacAI Updatejoymorrison10
 
Where to Stay in Lagos, Portugal.pptxasd
Where to Stay in Lagos, Portugal.pptxasdWhere to Stay in Lagos, Portugal.pptxasd
Where to Stay in Lagos, Portugal.pptxasdusmanghaniwixpatriot
 
69 Girls ✠ 9599264170 ✠ Call Girls In East Of Kailash (VIP)
69 Girls ✠ 9599264170 ✠ Call Girls In East Of Kailash (VIP)69 Girls ✠ 9599264170 ✠ Call Girls In East Of Kailash (VIP)
69 Girls ✠ 9599264170 ✠ Call Girls In East Of Kailash (VIP)Escort Service
 
(8264348440) 🔝 Call Girls In Nand Nagri 🔝 Delhi NCR
(8264348440) 🔝 Call Girls In Nand Nagri 🔝 Delhi NCR(8264348440) 🔝 Call Girls In Nand Nagri 🔝 Delhi NCR
(8264348440) 🔝 Call Girls In Nand Nagri 🔝 Delhi NCRsoniya singh
 
question 2: airplane vocabulary presentation
question 2: airplane vocabulary presentationquestion 2: airplane vocabulary presentation
question 2: airplane vocabulary presentationcaminantesdaauga
 
Dubai Call Girls O528786472 Call Girls Dubai Big Juicy
Dubai Call Girls O528786472 Call Girls Dubai Big JuicyDubai Call Girls O528786472 Call Girls Dubai Big Juicy
Dubai Call Girls O528786472 Call Girls Dubai Big Juicyhf8803863
 
8377087607 Full Enjoy @24/7 Call Girls in INA Market Dilli Hatt Delhi NCR
8377087607 Full Enjoy @24/7 Call Girls in INA Market Dilli Hatt Delhi NCR8377087607 Full Enjoy @24/7 Call Girls in INA Market Dilli Hatt Delhi NCR
8377087607 Full Enjoy @24/7 Call Girls in INA Market Dilli Hatt Delhi NCRdollysharma2066
 
How Safe Is It To Witness Whales In Maui’s Waters
How Safe Is It To Witness Whales In Maui’s WatersHow Safe Is It To Witness Whales In Maui’s Waters
How Safe Is It To Witness Whales In Maui’s WatersMakena Coast Charters
 
Hoi An Ancient Town, Vietnam (越南 會安古鎮).ppsx
Hoi An Ancient Town, Vietnam (越南 會安古鎮).ppsxHoi An Ancient Town, Vietnam (越南 會安古鎮).ppsx
Hoi An Ancient Town, Vietnam (越南 會安古鎮).ppsxChung Yen Chang
 
5S - House keeping (Seiri, Seiton, Seiso, Seiketsu, Shitsuke)
5S - House keeping (Seiri, Seiton, Seiso, Seiketsu, Shitsuke)5S - House keeping (Seiri, Seiton, Seiso, Seiketsu, Shitsuke)
5S - House keeping (Seiri, Seiton, Seiso, Seiketsu, Shitsuke)Mazie Garcia
 

Último (20)

"Fly with Ease: Booking Your Flights with Air Europa"
"Fly with Ease: Booking Your Flights with Air Europa""Fly with Ease: Booking Your Flights with Air Europa"
"Fly with Ease: Booking Your Flights with Air Europa"
 
Italia Lucca 1 Un tesoro nascosto tra le sue mura
Italia Lucca 1 Un tesoro nascosto tra le sue muraItalia Lucca 1 Un tesoro nascosto tra le sue mura
Italia Lucca 1 Un tesoro nascosto tra le sue mura
 
Apply Indian E-Visa Process Online (Evisa)
Apply Indian E-Visa Process Online (Evisa)Apply Indian E-Visa Process Online (Evisa)
Apply Indian E-Visa Process Online (Evisa)
 
Haitian culture and stuff and places and food and travel.pptx
Haitian culture and stuff and places and food and travel.pptxHaitian culture and stuff and places and food and travel.pptx
Haitian culture and stuff and places and food and travel.pptx
 
Authentic Travel Experience 2024 Greg DeShields.pptx
Authentic Travel Experience 2024 Greg DeShields.pptxAuthentic Travel Experience 2024 Greg DeShields.pptx
Authentic Travel Experience 2024 Greg DeShields.pptx
 
Exploring Sicily Your Comprehensive Ebook Travel Guide
Exploring Sicily Your Comprehensive Ebook Travel GuideExploring Sicily Your Comprehensive Ebook Travel Guide
Exploring Sicily Your Comprehensive Ebook Travel Guide
 
Inspirational Quotes About Italy and Food
Inspirational Quotes About Italy and FoodInspirational Quotes About Italy and Food
Inspirational Quotes About Italy and Food
 
Aeromexico Airlines Flight Name Change Policy
Aeromexico Airlines Flight Name Change PolicyAeromexico Airlines Flight Name Change Policy
Aeromexico Airlines Flight Name Change Policy
 
Revolutionalizing Travel: A VacAI Update
Revolutionalizing Travel: A VacAI UpdateRevolutionalizing Travel: A VacAI Update
Revolutionalizing Travel: A VacAI Update
 
Enjoy ➥8448380779▻ Call Girls In Sector 62 Noida Escorts Delhi NCR
Enjoy ➥8448380779▻ Call Girls In Sector 62 Noida Escorts Delhi NCREnjoy ➥8448380779▻ Call Girls In Sector 62 Noida Escorts Delhi NCR
Enjoy ➥8448380779▻ Call Girls In Sector 62 Noida Escorts Delhi NCR
 
Where to Stay in Lagos, Portugal.pptxasd
Where to Stay in Lagos, Portugal.pptxasdWhere to Stay in Lagos, Portugal.pptxasd
Where to Stay in Lagos, Portugal.pptxasd
 
69 Girls ✠ 9599264170 ✠ Call Girls In East Of Kailash (VIP)
69 Girls ✠ 9599264170 ✠ Call Girls In East Of Kailash (VIP)69 Girls ✠ 9599264170 ✠ Call Girls In East Of Kailash (VIP)
69 Girls ✠ 9599264170 ✠ Call Girls In East Of Kailash (VIP)
 
Enjoy ➥8448380779▻ Call Girls In Sector 74 Noida Escorts Delhi NCR
Enjoy ➥8448380779▻ Call Girls In Sector 74 Noida Escorts Delhi NCREnjoy ➥8448380779▻ Call Girls In Sector 74 Noida Escorts Delhi NCR
Enjoy ➥8448380779▻ Call Girls In Sector 74 Noida Escorts Delhi NCR
 
(8264348440) 🔝 Call Girls In Nand Nagri 🔝 Delhi NCR
(8264348440) 🔝 Call Girls In Nand Nagri 🔝 Delhi NCR(8264348440) 🔝 Call Girls In Nand Nagri 🔝 Delhi NCR
(8264348440) 🔝 Call Girls In Nand Nagri 🔝 Delhi NCR
 
question 2: airplane vocabulary presentation
question 2: airplane vocabulary presentationquestion 2: airplane vocabulary presentation
question 2: airplane vocabulary presentation
 
Dubai Call Girls O528786472 Call Girls Dubai Big Juicy
Dubai Call Girls O528786472 Call Girls Dubai Big JuicyDubai Call Girls O528786472 Call Girls Dubai Big Juicy
Dubai Call Girls O528786472 Call Girls Dubai Big Juicy
 
8377087607 Full Enjoy @24/7 Call Girls in INA Market Dilli Hatt Delhi NCR
8377087607 Full Enjoy @24/7 Call Girls in INA Market Dilli Hatt Delhi NCR8377087607 Full Enjoy @24/7 Call Girls in INA Market Dilli Hatt Delhi NCR
8377087607 Full Enjoy @24/7 Call Girls in INA Market Dilli Hatt Delhi NCR
 
How Safe Is It To Witness Whales In Maui’s Waters
How Safe Is It To Witness Whales In Maui’s WatersHow Safe Is It To Witness Whales In Maui’s Waters
How Safe Is It To Witness Whales In Maui’s Waters
 
Hoi An Ancient Town, Vietnam (越南 會安古鎮).ppsx
Hoi An Ancient Town, Vietnam (越南 會安古鎮).ppsxHoi An Ancient Town, Vietnam (越南 會安古鎮).ppsx
Hoi An Ancient Town, Vietnam (越南 會安古鎮).ppsx
 
5S - House keeping (Seiri, Seiton, Seiso, Seiketsu, Shitsuke)
5S - House keeping (Seiri, Seiton, Seiso, Seiketsu, Shitsuke)5S - House keeping (Seiri, Seiton, Seiso, Seiketsu, Shitsuke)
5S - House keeping (Seiri, Seiton, Seiso, Seiketsu, Shitsuke)
 

Context-Aware Points of Interest Suggestion with Dynamic Weather Data Management

  • 1. Context-Aware Points of Interest Suggestion with Dynamic Weather Data Management Matthias Braunhofer, Mehdi Elahi, Francesco Ricci, and Thomas Schievenin Faculty of Computer Science Free University of Bozen – Bolzano, Italy ENTER 2014 Research Track Slide Number 1
  • 2. Agenda • Recommender Systems and Context-Awareness • Related Work • Weather-Aware Recommendations • Experimental Evaluation • Conclusions and Future Work ENTER 2014 Research Track Slide Number 2
  • 3. Recommender Systems (RSs) • Goal: recommend new, relevant items to users based on their feedback on a sample of items (training set) – Explicit feedback (ratings) vs. implicit feedback (purchase / browsing history) • Two basic technical approaches: – Collaborative filtering (CF) – Content-based ENTER 2014 Research Track Slide Number 3
  • 4. Context is Essential • Main idea: users can experience items differently depending on the current contextual situation (e.g., season, weather, temperature, mood) • Example: ENTER 2014 Research Track Slide Number 4
  • 5. Context-Aware Recommender Systems (CARSs) • CARSs extend RSs beyond users and items to the contexts in which items are experienced by users – Rating prediction function is: R: Users x Items x Context  Ratings 1 3 2 3 4 5 3 1 5 2 5 2 4 5 ? 5 4 ? 4 2 3 3 4 3 3 ENTER 2014 Research Track 5 5 3 3 5 Slide Number 5
  • 6. Using Weather in RSs • Hypothesis: weather conditions at places of interest (POIs), together with past ratings for POIs under several distinct weather conditions, can be used to improve the choice satisfaction and perceived recommendation quality • Example: ? 5 1 ? ENTER 2014 Research Track Slide Number 6
  • 7. Agenda ENTER 2014 Research Track Slide Number 7
  • 8. Context-Aware Matrix Factorization (CAMF) • CAMF (Baltrunas et al., 2011) extends matrix factorization by incorporating baseline parameters for contextual conditions-item pairs, which capture the deviation of items‘ ratings due to context k ˆ ruic1 ,...,ck = i + bu + ∑ bic j + pu qiT j=1 • Main limitations: ī: average rating of item i bu: baseline for user u bicj: baseline for contextual condition cj and item i pu: latent factor vector of user u qi: latent factor vector of item i – Fails to produce personalized recommendations for new users – Uses the weather condition around the user rather than the weather conditions at POIs ENTER 2014 Research Track Slide Number 9
  • 9. Mobile CARSs (1/2) • liveCities (Martin et al., 2011): supports tourists by sending them push-based notifications when they enter a certain area and their context matches pre-defined conditions • Main limitations: - Considers only the temperature and not the weather Uses pre-defined recommendations rather than a predictive model Is only a research prototype ENTER 2014 Research Track Slide Number 10
  • 10. Mobile CARSs (2/2) • VISIT (Meehan et al., 2013): hybrid mobile RS that uses several contextual factors (i.e., location, time, weather, social media sentiment) to support tourist‘s decisionmaking process • Main limitation: - System is only a proposal and not yet implemented ENTER 2014 Research Track Slide Number 11
  • 11. Agenda ENTER 2014 Research Track Slide Number 12
  • 12. STS – South Tyrol Suggests ENTER 2014 Research Track Slide Number 13
  • 13. Weather-Aware Recommendations • Main idea: use CAMF (Baltrunas et al., 2011) as starting point and incorporate additional user attributes (i.e., gender, birth date and personality trait information) • Advantage: can produce personalized recommendations for users without ratings CAMF k ˆ ruic1 ,...,ck = i + bu + ∑ bic j + qiT ( pu + j=1 ∑ ya ) a∈A(u ) A distinct factor vector ya corresponds to each user attribute to describe a user through the set of user-associated attributes ENTER 2014 Research Track Slide Number 14
  • 14. Phase 1: Training • The system learns the parameters offline (i.e., once every 5 minutes) by minimizing the regularized squared error on the set of known ratings K (the training set): Prediction error minb*,q*, p*,y* ∑ 2 (u,i,c1 ,...,ck )∈K k ˆ (ruic1 ,...,ck − ruic1 ,...,ck ) + λ (b + ∑ b + qi + pu + 2 u j=1 2 ic j 2 2 ∑ 2 ya ) a∈A(u ) Regularization term to avoid overfitting • Minimization performed by stochastic gradient descent ENTER 2014 Research Track Slide Number 15
  • 15. Phase 2: Recommendation • Retrieve the weather and temperature values for the 116 municipalities of South Tyrol by querying Mondometeo – Note: This information can be cached • For each POI in the database: – Look up the POI‘s location – Assign the weather and temperature values retrieved for the closest municipality – Compute a rating prediction, considering the weather and temperature conditions along with other known contextual conditions • Recommend the top-20 POIs to the user ENTER 2014 Research Track Slide Number 16
  • 16. Agenda ENTER 2014 Research Track Slide Number 17
  • 17. Experimental Methodology • Live user study where our proposed weather-aware system (STS) was compared with a variant (STS-S) that has the same graphical UI but does not use the weather context when generating recommendations • We have designed a specific user task and used a questionnaire for assessing the perceived recommendation quality and choice satisfaction (Knijnenburg et al., 2012) • 54 subjects that were randomly divided in two equal groups assigned to STS and STS-S (27 each) ENTER 2014 Research Track Slide Number 18
  • 18. User Task (1/2) • Users were supposed to: – have an afternoon off and to look for attractions / events in South Tyrol – consider the contextual conditions relevant for them and to specify them in the system settings – browse the attractions / events sections and check whether they could find something interesting for them – browse the system suggestions (recommendations), and select and bookmark the one that they believe fits their needs and wants – fill up a survey (Knijnenburg et al., 2012), which measures perceived recommendation quality and choice satisfaction ENTER 2014 Research Track Slide Number 19
  • 19. User Task (2/2) • After this initial interaction, users had the opportunity to double check the weather conditions at the bookmarked POI by accessing on a computer the Mondometeo website – Will this knowledge influence the users choice? – Hypothesis: STS has exploited this information so the user should have already chosen items that are "compatible" with the weather • Users were then asked whether they wanted to change their preferred POI and bookmark another one, i.e., if they believed that, because of the weather conditions at the bookmarked POI, their previous choice was not anymore considered to be appropriate ENTER 2014 Research Track Slide Number 20
  • 20. Results (1/2) Number of unsatisfied users, i.e., those that changed their bookmarked POI after having double checked the weather conditions at the POI ENTER 2014 Research Track Slide Number 21
  • 21. Results (2/2) Choice satisfaction Perceived rec. quality Statement STS avg. STS-S avg. pvalue 1. I liked the items suggested by the system 4.0 3.7 0.20 2. The suggested items fitted my preference. 3.4 3.4 0.56 3. The suggested items were well-chosen 4. The suggested items were relevant. 3.5 3.7 3.3 3.2 0.13 0.04 5. The system suggested too many bad items. 2.9 2.7 6. I didn’t like any of the suggested items. 3.8 3.3 0.14 < 0.001 7. The items I selected were “the best among the worst”. 8. I like the item I’ve chosen. 3.1 4.6 2.8 4.3 0.20 0.02 9. I was excited about my chosen item. 4.0 3.7 0.03 10. I enjoyed watching my chosen item. 3.7 3.9 0.79 11. The items I watched were a waste of my time. 3.5 3.5 0.42 12. The chosen item fitted my preference. 3.9 4.0 0.71 ENTER 2014 Research Track Slide Number 22
  • 22. Agenda ENTER 2014 Research Track Slide Number 23
  • 23. Conclusions • STS = mobile CARS that recommends POIs using a set of contextual factors that include the current weather conditions at the recommended POIs • Novelty is the usage of up-to-date weather data into a matrix factorization algorithm to generate personalized context-aware recommendations • Usage of weather data significantly improves the users‘ perceived recommendation quality and choice satisfaction ENTER 2014 Research Track Slide Number 24
  • 24. Future Work • Extended analysis – To better understand potential performance differences among the compared CARSs, which may be due to different usage of the weather contextual factors – To test our proposed weather-aware CARS with a larger number of users and a larger rating dataset ENTER 2014 Research Track Slide Number 25
  • 25. Questions? Thank you. ENTER 2014 Research Track Slide Number 26

Notas do Editor

  1. ?