SlideShare a Scribd company logo
1 of 1
Download to read offline
Predicting User Engagement in Twitter with Collaborative Ranking 
Ernesto Diaz-Aviles∗, Hoang Thanh Lam, Fabio Pinelli, Stefano Braghin, 
Yiannis Gkoufas, Michele Berlingerio, and Francesco Calabrese 
IBM Research – Ireland 
Problem 
▶ Current methods of Collaborative Filtering (CF) evaluation: (i) quality of a predicted rating or 
(ii) the ranking performance for top-n recommended items 
▶ These evaluation methods are rather limiting and neglect other dimensions that could better 
characterize a well-perceived recommendation 
▶ The task in this work: predict which items generate the highest user engagement 
Contribution: 
Collaborative ranking approach for user engagement prediction in Twitter 
Twitter 
IMDb 
   
  
(1) Extract features 
(2) Learn ranking function for 
user engagement prediction 
 
 
 
Feature Extraction 
▶ User rating, F1 = ruid 
▶ Deviation of user rating from the median of previous user ratings, 
i.e., F2 = ruid − ˜ru 
▶ Average user engagement from her history, i.e., 
F3 = engagement(u)0.5 
▶ F4 = [engagement(u)  0] , where [a] is one if a is true and 0 if 
a is false 
▶ Average rating per user, F5 = ¯ ru 
▶ F6 = 
( #friends(u) 
#followers(u) 
)0.5 
▶ User tweet count: F7 = #tweets(u)0.5 
▶ Average user engagement for a given movie item i , i.e., 
F8 = engagement(i)0.5 
▶ Boolean indicator that takes the value of 1 if the average user 
engagement for item i is greater than 0, and 0 otherwise, i.e., 
F9 = [engagement(i)  0] 
▶ Average rating per item, F10 = ¯ri 
▶ Average ratio of number of user friends to the number of her 
followers aggregated over the movie item i: 
F11 = 
(1 
K 
ΣKu 
∈Ui 
#friends(u) 
#followers(u) 
)0.5 
where 
Ui := {u ∈ U|(u, j, d) ∈ S ∧ j = i} and K = |Ui | . 
▶ Average of user tweet counts aggregated per item i : 
F12 = 
(1 
K 
ΣKu 
∈Ui #tweets(u) 
)0.5 
▶ User mentions: F13 = [has mention(d)] 
▶ Tweet is a retweet (retweeted status): F14 = [is retweet(d)] 
▶ The same field retweeted status for d also includes the tweet 
id (tweet id(do)) of the original tweet, if such tweet do is present in 
the dataset we know that it received a non-zero engagement, F15 
represents this additional information for do: 
F15ϕ(u,i,do) = [is retweet of (d, do)] , 
where is retweet of (d, do) is true if d is do’s retweet. 
▶ The frequency of observed engagement (i.e., retweet count) 
extracted per item from the retweeted status field: 
F16 = Σ 
d∈is retweet(D) 
[engagement(i)] . 
Collaborative Ranking for User Engagement 
▶ Procedure: CRUE 
Input: Training set S = {ϕ(u, i, d)k, yuidk}mk 
=1 
Output: Ranking function f (u, i, d, Θ) 
1: Scale and normalize feature vectors ϕ(u, i, d) 
2: Remove user outliers 
3: Learn a ranking function 
f (u, d, i, Θ) = h(ϕ(u, i, d); θ) + buid 
by optimizing nDCG@10 directly 
4: return f (u, i, d, Θ) 
Ranking Function 
▶ Linear Ensemble: LambdaMART + MART 
Results 
Recommender nDCG@10 
CRUE (our approach) 0.8701 
FM 0.8023 
recRating 0.8182 
recHEI 0.8031 
recRandom 0.7532 
Conclusion 
▶ Our collaborative ranking approach is able to predict 
user engagement in Twitter 
▶ Twitter rich metadata + explicit feedback (i.e., rating) → high-quality feature vectors 
▶ Effective and on-the-fly prediction → explicit feedback 
(rating) + the historic user engagement per item 
*Corresponding author: Ernesto Diaz-Aviles e.diaz-aviles@ie.ibm.com

More Related Content

Recently uploaded

Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptx
Cherry
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Sérgio Sacani
 
development of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusdevelopment of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virus
NazaninKarimi6
 
CYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptx
Cherry
 
Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.
Cherry
 
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cherry
 

Recently uploaded (20)

Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptx
 
Genome organization in virus,bacteria and eukaryotes.pptx
Genome organization in virus,bacteria and eukaryotes.pptxGenome organization in virus,bacteria and eukaryotes.pptx
Genome organization in virus,bacteria and eukaryotes.pptx
 
Genetics and epigenetics of ADHD and comorbid conditions
Genetics and epigenetics of ADHD and comorbid conditionsGenetics and epigenetics of ADHD and comorbid conditions
Genetics and epigenetics of ADHD and comorbid conditions
 
Cyanide resistant respiration pathway.pptx
Cyanide resistant respiration pathway.pptxCyanide resistant respiration pathway.pptx
Cyanide resistant respiration pathway.pptx
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
 
CURRENT SCENARIO OF POULTRY PRODUCTION IN INDIA
CURRENT SCENARIO OF POULTRY PRODUCTION IN INDIACURRENT SCENARIO OF POULTRY PRODUCTION IN INDIA
CURRENT SCENARIO OF POULTRY PRODUCTION IN INDIA
 
Cot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNACot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNA
 
Site Acceptance Test .
Site Acceptance Test                    .Site Acceptance Test                    .
Site Acceptance Test .
 
Role of AI in seed science Predictive modelling and Beyond.pptx
Role of AI in seed science  Predictive modelling and  Beyond.pptxRole of AI in seed science  Predictive modelling and  Beyond.pptx
Role of AI in seed science Predictive modelling and Beyond.pptx
 
Dr. E. Muralinath_ Blood indices_clinical aspects
Dr. E. Muralinath_ Blood indices_clinical  aspectsDr. E. Muralinath_ Blood indices_clinical  aspects
Dr. E. Muralinath_ Blood indices_clinical aspects
 
Use of mutants in understanding seedling development.pptx
Use of mutants in understanding seedling development.pptxUse of mutants in understanding seedling development.pptx
Use of mutants in understanding seedling development.pptx
 
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot Girls
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot GirlsKanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot Girls
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot Girls
 
Clean In Place(CIP).pptx .
Clean In Place(CIP).pptx                 .Clean In Place(CIP).pptx                 .
Clean In Place(CIP).pptx .
 
development of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusdevelopment of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virus
 
CYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptx
 
Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
 
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRLGwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
 
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
 
Plasmid: types, structure and functions.
Plasmid: types, structure and functions.Plasmid: types, structure and functions.
Plasmid: types, structure and functions.
 

Featured

Social Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsSocial Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie Insights
Kurio // The Social Media Age(ncy)
 

Featured (20)

AI Trends in Creative Operations 2024 by Artwork Flow.pdf
AI Trends in Creative Operations 2024 by Artwork Flow.pdfAI Trends in Creative Operations 2024 by Artwork Flow.pdf
AI Trends in Creative Operations 2024 by Artwork Flow.pdf
 
Skeleton Culture Code
Skeleton Culture CodeSkeleton Culture Code
Skeleton Culture Code
 
PEPSICO Presentation to CAGNY Conference Feb 2024
PEPSICO Presentation to CAGNY Conference Feb 2024PEPSICO Presentation to CAGNY Conference Feb 2024
PEPSICO Presentation to CAGNY Conference Feb 2024
 
Content Methodology: A Best Practices Report (Webinar)
Content Methodology: A Best Practices Report (Webinar)Content Methodology: A Best Practices Report (Webinar)
Content Methodology: A Best Practices Report (Webinar)
 
How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024
 
Social Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsSocial Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie Insights
 
Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024
 
5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary
 
ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd
 
Getting into the tech field. what next
Getting into the tech field. what next Getting into the tech field. what next
Getting into the tech field. what next
 
Google's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search IntentGoogle's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search Intent
 
How to have difficult conversations
How to have difficult conversations How to have difficult conversations
How to have difficult conversations
 
Introduction to Data Science
Introduction to Data ScienceIntroduction to Data Science
Introduction to Data Science
 
Time Management & Productivity - Best Practices
Time Management & Productivity -  Best PracticesTime Management & Productivity -  Best Practices
Time Management & Productivity - Best Practices
 
The six step guide to practical project management
The six step guide to practical project managementThe six step guide to practical project management
The six step guide to practical project management
 
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
 
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
 
12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work
 
ChatGPT webinar slides
ChatGPT webinar slidesChatGPT webinar slides
ChatGPT webinar slides
 
More than Just Lines on a Map: Best Practices for U.S Bike Routes
More than Just Lines on a Map: Best Practices for U.S Bike RoutesMore than Just Lines on a Map: Best Practices for U.S Bike Routes
More than Just Lines on a Map: Best Practices for U.S Bike Routes
 

Predicting User Engagement in Twitter with Collaborative Ranking. IBM Research - Ireland @ RecSys Challenge 2014 (3rd Place Winners)

  • 1. Predicting User Engagement in Twitter with Collaborative Ranking Ernesto Diaz-Aviles∗, Hoang Thanh Lam, Fabio Pinelli, Stefano Braghin, Yiannis Gkoufas, Michele Berlingerio, and Francesco Calabrese IBM Research – Ireland Problem ▶ Current methods of Collaborative Filtering (CF) evaluation: (i) quality of a predicted rating or (ii) the ranking performance for top-n recommended items ▶ These evaluation methods are rather limiting and neglect other dimensions that could better characterize a well-perceived recommendation ▶ The task in this work: predict which items generate the highest user engagement Contribution: Collaborative ranking approach for user engagement prediction in Twitter Twitter IMDb      (1) Extract features (2) Learn ranking function for user engagement prediction    Feature Extraction ▶ User rating, F1 = ruid ▶ Deviation of user rating from the median of previous user ratings, i.e., F2 = ruid − ˜ru ▶ Average user engagement from her history, i.e., F3 = engagement(u)0.5 ▶ F4 = [engagement(u) 0] , where [a] is one if a is true and 0 if a is false ▶ Average rating per user, F5 = ¯ ru ▶ F6 = ( #friends(u) #followers(u) )0.5 ▶ User tweet count: F7 = #tweets(u)0.5 ▶ Average user engagement for a given movie item i , i.e., F8 = engagement(i)0.5 ▶ Boolean indicator that takes the value of 1 if the average user engagement for item i is greater than 0, and 0 otherwise, i.e., F9 = [engagement(i) 0] ▶ Average rating per item, F10 = ¯ri ▶ Average ratio of number of user friends to the number of her followers aggregated over the movie item i: F11 = (1 K ΣKu ∈Ui #friends(u) #followers(u) )0.5 where Ui := {u ∈ U|(u, j, d) ∈ S ∧ j = i} and K = |Ui | . ▶ Average of user tweet counts aggregated per item i : F12 = (1 K ΣKu ∈Ui #tweets(u) )0.5 ▶ User mentions: F13 = [has mention(d)] ▶ Tweet is a retweet (retweeted status): F14 = [is retweet(d)] ▶ The same field retweeted status for d also includes the tweet id (tweet id(do)) of the original tweet, if such tweet do is present in the dataset we know that it received a non-zero engagement, F15 represents this additional information for do: F15ϕ(u,i,do) = [is retweet of (d, do)] , where is retweet of (d, do) is true if d is do’s retweet. ▶ The frequency of observed engagement (i.e., retweet count) extracted per item from the retweeted status field: F16 = Σ d∈is retweet(D) [engagement(i)] . Collaborative Ranking for User Engagement ▶ Procedure: CRUE Input: Training set S = {ϕ(u, i, d)k, yuidk}mk =1 Output: Ranking function f (u, i, d, Θ) 1: Scale and normalize feature vectors ϕ(u, i, d) 2: Remove user outliers 3: Learn a ranking function f (u, d, i, Θ) = h(ϕ(u, i, d); θ) + buid by optimizing nDCG@10 directly 4: return f (u, i, d, Θ) Ranking Function ▶ Linear Ensemble: LambdaMART + MART Results Recommender nDCG@10 CRUE (our approach) 0.8701 FM 0.8023 recRating 0.8182 recHEI 0.8031 recRandom 0.7532 Conclusion ▶ Our collaborative ranking approach is able to predict user engagement in Twitter ▶ Twitter rich metadata + explicit feedback (i.e., rating) → high-quality feature vectors ▶ Effective and on-the-fly prediction → explicit feedback (rating) + the historic user engagement per item *Corresponding author: Ernesto Diaz-Aviles e.diaz-aviles@ie.ibm.com