SlideShare una empresa de Scribd logo
1 de 17
Descargar para leer sin conexión
Graph Pattern Entity Ranking Model
for Knowledge Graph Completion
Takuma Ebisu @Sokendai, Ryutaro Ichise @NIIC
NAACL 2019
Introduced by Naomi Shiraishi
January 29th, 2020
“Graph Pattern Entity Ranking Model for Knowledge Graph
Completion”
• Why this work?
• Knowledge graph is important to deal lots of AI tasks.
• But lots of missing facts (=relations, edges) exist -> Link prediction task
• Many Knowledge Graph Embedding models were developed to tackle the task,
however…
• Black box! User has no idea how the information is processed.
• Model is difficult to interpret.
• Approach
• Use Graph Patterns in knowledge graph
• GPro: Use conditional probability
• GRank: Use Graph Pattern Association Rues as ranking model of entities
• Link Prediction で SOTA!
• Easy to implement
• Validation
• Dataset: WN18, FB15k, WN18RR, FB15k-237
• Metrics: MRR, HITS@n
• Proposed by Fan et al. (2015) for social network graph
• This paper proposes GPARs for Knowledge Graph
Preparation: Graph Pattern Association Rues (GPARs)
𝑡
ℎ
𝑟
Preparation: Graph Pattern Association Rues (GPARs)
Graph Pattern:
Graph Pattern Association Rues:
Example
Set of variables
Set of Relations
Located_in
Preparation: Queries and answers
• Devide a knowledge graph G into queries and answers to use as
training data for the model
Training queries
The answers of training queries
?
ℎ
𝑟
𝑡
𝑟
?
Preparation: Graph Pattern Matching Function
• A matching function of 𝐺𝑃(',)) on ℎ, 𝑡 ∈ 𝐸 × 𝐸 is
an injective function(単射関数) m: 𝑉12 → 𝐸
• Matching function satisfies:
• : set of all matching functions
Example
matching function of
Link prediction 1: Graph Pattern Probability Model (GPro)
• Use 2 confidence values (confidence for tail and head)
Conditional Probability
h を通るGraph Patternの先の候補の数
hを通るGraph Patternの先が
正しときのentityの数
Link prediction 1: Graph Pattern Probability Model (GPro)
Example
𝐴𝑅6(',)) = { 𝑧, member_of, 𝑥 , (𝑧, nafonality, 𝑦)} → (𝑥, 𝑙𝑜𝑐𝑎𝑡𝑒𝑑_𝑖𝑛, 𝑦)
𝐴𝑅F(',)) = { 𝑧, manager_of, 𝑥 , (𝑧, nafonality, 𝑦)} → (𝑥, 𝑙𝑜𝑐𝑎𝑡𝑒𝑑_𝑖𝑛, 𝑦)
𝐶𝑜𝑛𝑓IJKL(𝐴𝑅6(',))) =
(𝑈. 𝐾. , 𝐺𝑒𝑟𝑚𝑎𝑛𝑦)
(𝑈. 𝐾. , 𝐹𝑟𝑎𝑛𝑐𝑒, 𝐼𝑡𝑎𝑙𝑦, 𝑈. 𝐾. , 𝐺𝑒𝑟𝑚𝑎𝑛𝑦)
=
2
5
𝐶𝑜𝑛𝑓IJKL(𝐴𝑅F(',))) =
(𝐺𝑒𝑟𝑚𝑎𝑛𝑦)
(𝐹𝑟𝑎𝑛𝑐𝑒, 𝐺𝑒𝑟𝑚𝑎𝑛𝑦)
=
1
2
𝐴𝑅6(',)) is underestimated L
Link prediction 2: GRank
• Consider GPARs as entity ranking system
• Define a scoring function whose arguments are Graph Pattern and a
pair of entities:
• When applying a GPARs to answer a query, create rankings of the
candidate entities by their score
• We can evaluate GPAR as an entity ranking system by evaluating
output rankings.
Link prediction 2: GRank
• Distributed Ranking
• Given a pattern GP and a query (h, r, ?), we obtain distributed
rankings of entities according to their scores:
𝑑𝑟𝑎𝑛𝑘K,W = X
6
Y
0
When 𝑎 + 1 ≤ 𝑖 ≤ 𝑎 + 𝑏
Otherwise
Entity j
rank i
1
2
3
…
a: スコアがentity j より⼤きいentityの数
b: スコアがentity j と同じentity の数
Link prediction 2: GRank
• To evaluate GPAR, define metric to evaluate distributed rankings of
entities by generalizing the average precision
• For a pattern GP and a training query (h, r, ?), the distributed
precision at rank k is defined:
Entity j
rank i
1
2
…
…
rank k
Link prediction 2: GRank
• For a pattern GP and a training query (h, r, ?), the distributed average
precision is defined:
• The distributed mean average precision
for a GPAR GP->r is defined:
Entity j
k
1
2
3
…
Experiments: Baselines
• Knowledge graph embedding models
• Translation-based: TransE (Bordes et al. 2013), TrousE (Ebisu and Ichise, 2018)
• Simple (h+r = t) and effective
• Bilinear: RESCAL (Nickel et al., 2011)
• Neural network-based: GCNs (Duvenaud et al., 2015)
• Observed Feature models:
predict based on observable features in the graph (???)
• Node+LinkFeat (Toutanova et al., 2015)
• only use on-hop information
• PRA (Lao and Cohen, 2010)
• use multi-hop information
Results
• WN18 from WordNet (1995): Well constructed. Few missing or wrong facts
• FB15k from Freebase (2008): Many missing facts.
• WN18 and FB15k have redundancy in the form of reverse relations
-> WN18RR and FB15k-237: the inverse relation of other relations are removed
• Graph pattern is restricted to connected and closed (No branch between x and y)
• Limit the size of GP to be <= 3
Mean reciprocal rank
Observed
feature models
The proportion of test queries whose corresponding
entities are ranked in the top n of the obtained rankings
Knowledge graph
embedding models
eliminating entities
whose
corresponding
triples (except the
target triple) were
included in the
training dataset.
Results
• Example of antecedent patterns for dMAP_tail which were given high
ranks by GRank for FB15k
high Low Low
This only works when an
individual has more
than two siblings
Individual’s parents are
often missing in FB15k
Conclusions
• Defined GPAR for a knowledge graph.
• Defined the standard confidence measures of GPARs for the link
prediction.
• Introduced GPro model
• Introduced GRank model using distributed ranking -> SOTA
• Future work: Extend GRank to use more complex pattern
• Graph pattern size > 3
論文紹介:Graph Pattern Entity Ranking Model for Knowledge Graph Completion

Más contenido relacionado

La actualidad más candente

Graph based Semi Supervised Learning V1
Graph based Semi Supervised Learning V1Graph based Semi Supervised Learning V1
Graph based Semi Supervised Learning V1
Neeta Pande
 
presentation
presentationpresentation
presentation
jie ren
 
AllegroGraph - Cognitive Probability Graph webcast
AllegroGraph - Cognitive Probability Graph webcastAllegroGraph - Cognitive Probability Graph webcast
AllegroGraph - Cognitive Probability Graph webcast
Franz Inc. - AllegroGraph
 

La actualidad más candente (19)

Graph based Semi Supervised Learning V1
Graph based Semi Supervised Learning V1Graph based Semi Supervised Learning V1
Graph based Semi Supervised Learning V1
 
Power of Polyglot Search
Power of Polyglot SearchPower of Polyglot Search
Power of Polyglot Search
 
presentation
presentationpresentation
presentation
 
Ranking and Diversity in Recommendations - RecSys Stammtisch at SoundCloud, B...
Ranking and Diversity in Recommendations - RecSys Stammtisch at SoundCloud, B...Ranking and Diversity in Recommendations - RecSys Stammtisch at SoundCloud, B...
Ranking and Diversity in Recommendations - RecSys Stammtisch at SoundCloud, B...
 
Data Structures and Algorithm - Week 9 - Search Algorithms
Data Structures and Algorithm - Week 9 - Search AlgorithmsData Structures and Algorithm - Week 9 - Search Algorithms
Data Structures and Algorithm - Week 9 - Search Algorithms
 
Data visualization using R
Data visualization using RData visualization using R
Data visualization using R
 
A Graph-based Model for multimodal Information Retrieval (partially presented)
A Graph-based Model for multimodal Information Retrieval (partially presented)A Graph-based Model for multimodal Information Retrieval (partially presented)
A Graph-based Model for multimodal Information Retrieval (partially presented)
 
An Empirical Comparison of Knowledge Graph Embeddings for Item Recommendation
An Empirical Comparison of Knowledge Graph Embeddings for Item RecommendationAn Empirical Comparison of Knowledge Graph Embeddings for Item Recommendation
An Empirical Comparison of Knowledge Graph Embeddings for Item Recommendation
 
Big Data Graph Analytics
Big Data Graph AnalyticsBig Data Graph Analytics
Big Data Graph Analytics
 
Ikdd co ds2017presentation_v2
Ikdd co ds2017presentation_v2Ikdd co ds2017presentation_v2
Ikdd co ds2017presentation_v2
 
Data Structures and Algorithm - Week 8 - Minimum Spanning Trees
Data Structures and Algorithm - Week 8 - Minimum Spanning TreesData Structures and Algorithm - Week 8 - Minimum Spanning Trees
Data Structures and Algorithm - Week 8 - Minimum Spanning Trees
 
AllegroGraph - Cognitive Probability Graph webcast
AllegroGraph - Cognitive Probability Graph webcastAllegroGraph - Cognitive Probability Graph webcast
AllegroGraph - Cognitive Probability Graph webcast
 
Data Structures and Algorithm - Week 11 - Algorithm Analysis
Data Structures and Algorithm - Week 11 - Algorithm AnalysisData Structures and Algorithm - Week 11 - Algorithm Analysis
Data Structures and Algorithm - Week 11 - Algorithm Analysis
 
Bring survey sampling techniques into big data
Bring survey sampling techniques into big dataBring survey sampling techniques into big data
Bring survey sampling techniques into big data
 
Data Structures and Algorithm - Week 4 - Trees, Binary Trees
Data Structures and Algorithm - Week 4 - Trees, Binary TreesData Structures and Algorithm - Week 4 - Trees, Binary Trees
Data Structures and Algorithm - Week 4 - Trees, Binary Trees
 
Data Structures and Algorithm - Week 3 - Stacks and Queues
Data Structures and Algorithm - Week 3 - Stacks and QueuesData Structures and Algorithm - Week 3 - Stacks and Queues
Data Structures and Algorithm - Week 3 - Stacks and Queues
 
001 20151005 ranking_nodesingrowingnetwork
001 20151005 ranking_nodesingrowingnetwork001 20151005 ranking_nodesingrowingnetwork
001 20151005 ranking_nodesingrowingnetwork
 
Statistics in Data Science with Python
Statistics in Data Science with PythonStatistics in Data Science with Python
Statistics in Data Science with Python
 
AutoML - The Future of AI
AutoML - The Future of AIAutoML - The Future of AI
AutoML - The Future of AI
 

Similar a 論文紹介:Graph Pattern Entity Ranking Model for Knowledge Graph Completion

Big Data Conference
Big Data ConferenceBig Data Conference
Big Data Conference
DataTactics
 

Similar a 論文紹介:Graph Pattern Entity Ranking Model for Knowledge Graph Completion (20)

Synthetic Data Generation using exponential random Graph modeling
Synthetic Data Generation using exponential random Graph modelingSynthetic Data Generation using exponential random Graph modeling
Synthetic Data Generation using exponential random Graph modeling
 
Leveraging Graphs for Better AI
Leveraging Graphs for Better AILeveraging Graphs for Better AI
Leveraging Graphs for Better AI
 
Graph Analysis over Relational Database. Roberto Franchini - Arcade Analytics
Graph Analysis over Relational Database. Roberto Franchini - Arcade AnalyticsGraph Analysis over Relational Database. Roberto Franchini - Arcade Analytics
Graph Analysis over Relational Database. Roberto Franchini - Arcade Analytics
 
Final Defence.pptxFinal Defence.pptxFinal Defence.pptxFinal Defence.pptxFinal...
Final Defence.pptxFinal Defence.pptxFinal Defence.pptxFinal Defence.pptxFinal...Final Defence.pptxFinal Defence.pptxFinal Defence.pptxFinal Defence.pptxFinal...
Final Defence.pptxFinal Defence.pptxFinal Defence.pptxFinal Defence.pptxFinal...
 
Graph technology meetup slides
Graph technology meetup slidesGraph technology meetup slides
Graph technology meetup slides
 
Data Science as a Career and Intro to R
Data Science as a Career and Intro to RData Science as a Career and Intro to R
Data Science as a Career and Intro to R
 
What Makes Graph Queries Difficult?
What Makes Graph Queries Difficult?What Makes Graph Queries Difficult?
What Makes Graph Queries Difficult?
 
Graph analysis over relational database
Graph analysis over relational databaseGraph analysis over relational database
Graph analysis over relational database
 
STIC-D: algorithmic techniques for efficient parallel pagerank computation on...
STIC-D: algorithmic techniques for efficient parallel pagerank computation on...STIC-D: algorithmic techniques for efficient parallel pagerank computation on...
STIC-D: algorithmic techniques for efficient parallel pagerank computation on...
 
Benchmarking tool for graph algorithms
Benchmarking tool for graph algorithmsBenchmarking tool for graph algorithms
Benchmarking tool for graph algorithms
 
Leveraging Graphs for Better AI
Leveraging Graphs for Better AILeveraging Graphs for Better AI
Leveraging Graphs for Better AI
 
GraphGen: Conducting Graph Analytics over Relational Databases
GraphGen: Conducting Graph Analytics over Relational DatabasesGraphGen: Conducting Graph Analytics over Relational Databases
GraphGen: Conducting Graph Analytics over Relational Databases
 
GraphGen: Conducting Graph Analytics over Relational Databases
GraphGen: Conducting Graph Analytics over Relational DatabasesGraphGen: Conducting Graph Analytics over Relational Databases
GraphGen: Conducting Graph Analytics over Relational Databases
 
Benchmarking Tool for Graph Algorithms
Benchmarking Tool for Graph AlgorithmsBenchmarking Tool for Graph Algorithms
Benchmarking Tool for Graph Algorithms
 
A Blended Approach to Analytics at Data Tactics Corporation
A Blended Approach to Analytics at Data Tactics CorporationA Blended Approach to Analytics at Data Tactics Corporation
A Blended Approach to Analytics at Data Tactics Corporation
 
Big Data Conference
Big Data ConferenceBig Data Conference
Big Data Conference
 
Compressing Graphs and Indexes with Recursive Graph Bisection
Compressing Graphs and Indexes with Recursive Graph Bisection Compressing Graphs and Indexes with Recursive Graph Bisection
Compressing Graphs and Indexes with Recursive Graph Bisection
 
AnzoGraph DB: Driving AI and Machine Insights with Knowledge Graphs in a Conn...
AnzoGraph DB: Driving AI and Machine Insights with Knowledge Graphs in a Conn...AnzoGraph DB: Driving AI and Machine Insights with Knowledge Graphs in a Conn...
AnzoGraph DB: Driving AI and Machine Insights with Knowledge Graphs in a Conn...
 
Data analytics in computer networking
Data analytics in computer networkingData analytics in computer networking
Data analytics in computer networking
 
Improving Machine Learning using Graph Algorithms
Improving Machine Learning using Graph AlgorithmsImproving Machine Learning using Graph Algorithms
Improving Machine Learning using Graph Algorithms
 

Último

Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak HamilCara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Kandungan 087776558899
 
VIP Call Girls Palanpur 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Palanpur 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Palanpur 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Palanpur 7001035870 Whatsapp Number, 24/07 Booking
dharasingh5698
 

Último (20)

ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdfONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
 
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete RecordCCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
 
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
 
UNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its PerformanceUNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its Performance
 
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak HamilCara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
 
chapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineeringchapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineering
 
Double Revolving field theory-how the rotor develops torque
Double Revolving field theory-how the rotor develops torqueDouble Revolving field theory-how the rotor develops torque
Double Revolving field theory-how the rotor develops torque
 
Water Industry Process Automation & Control Monthly - April 2024
Water Industry Process Automation & Control Monthly - April 2024Water Industry Process Automation & Control Monthly - April 2024
Water Industry Process Automation & Control Monthly - April 2024
 
NFPA 5000 2024 standard .
NFPA 5000 2024 standard                                  .NFPA 5000 2024 standard                                  .
NFPA 5000 2024 standard .
 
VIP Call Girls Palanpur 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Palanpur 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Palanpur 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Palanpur 7001035870 Whatsapp Number, 24/07 Booking
 
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank  Design by Working Stress - IS Method.pdfIntze Overhead Water Tank  Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
 
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
 
data_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfdata_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdf
 
University management System project report..pdf
University management System project report..pdfUniversity management System project report..pdf
University management System project report..pdf
 
Unit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdfUnit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdf
 
Thermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.pptThermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.ppt
 
Design For Accessibility: Getting it right from the start
Design For Accessibility: Getting it right from the startDesign For Accessibility: Getting it right from the start
Design For Accessibility: Getting it right from the start
 
Work-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxWork-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptx
 
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
 
Generative AI or GenAI technology based PPT
Generative AI or GenAI technology based PPTGenerative AI or GenAI technology based PPT
Generative AI or GenAI technology based PPT
 

論文紹介:Graph Pattern Entity Ranking Model for Knowledge Graph Completion

  • 1. Graph Pattern Entity Ranking Model for Knowledge Graph Completion Takuma Ebisu @Sokendai, Ryutaro Ichise @NIIC NAACL 2019 Introduced by Naomi Shiraishi January 29th, 2020
  • 2. “Graph Pattern Entity Ranking Model for Knowledge Graph Completion” • Why this work? • Knowledge graph is important to deal lots of AI tasks. • But lots of missing facts (=relations, edges) exist -> Link prediction task • Many Knowledge Graph Embedding models were developed to tackle the task, however… • Black box! User has no idea how the information is processed. • Model is difficult to interpret. • Approach • Use Graph Patterns in knowledge graph • GPro: Use conditional probability • GRank: Use Graph Pattern Association Rues as ranking model of entities • Link Prediction で SOTA! • Easy to implement • Validation • Dataset: WN18, FB15k, WN18RR, FB15k-237 • Metrics: MRR, HITS@n
  • 3. • Proposed by Fan et al. (2015) for social network graph • This paper proposes GPARs for Knowledge Graph Preparation: Graph Pattern Association Rues (GPARs) 𝑡 ℎ 𝑟
  • 4. Preparation: Graph Pattern Association Rues (GPARs) Graph Pattern: Graph Pattern Association Rues: Example Set of variables Set of Relations Located_in
  • 5. Preparation: Queries and answers • Devide a knowledge graph G into queries and answers to use as training data for the model Training queries The answers of training queries ? ℎ 𝑟 𝑡 𝑟 ?
  • 6. Preparation: Graph Pattern Matching Function • A matching function of 𝐺𝑃(',)) on ℎ, 𝑡 ∈ 𝐸 × 𝐸 is an injective function(単射関数) m: 𝑉12 → 𝐸 • Matching function satisfies: • : set of all matching functions Example matching function of
  • 7. Link prediction 1: Graph Pattern Probability Model (GPro) • Use 2 confidence values (confidence for tail and head) Conditional Probability h を通るGraph Patternの先の候補の数 hを通るGraph Patternの先が 正しときのentityの数
  • 8. Link prediction 1: Graph Pattern Probability Model (GPro) Example 𝐴𝑅6(',)) = { 𝑧, member_of, 𝑥 , (𝑧, nafonality, 𝑦)} → (𝑥, 𝑙𝑜𝑐𝑎𝑡𝑒𝑑_𝑖𝑛, 𝑦) 𝐴𝑅F(',)) = { 𝑧, manager_of, 𝑥 , (𝑧, nafonality, 𝑦)} → (𝑥, 𝑙𝑜𝑐𝑎𝑡𝑒𝑑_𝑖𝑛, 𝑦) 𝐶𝑜𝑛𝑓IJKL(𝐴𝑅6(',))) = (𝑈. 𝐾. , 𝐺𝑒𝑟𝑚𝑎𝑛𝑦) (𝑈. 𝐾. , 𝐹𝑟𝑎𝑛𝑐𝑒, 𝐼𝑡𝑎𝑙𝑦, 𝑈. 𝐾. , 𝐺𝑒𝑟𝑚𝑎𝑛𝑦) = 2 5 𝐶𝑜𝑛𝑓IJKL(𝐴𝑅F(',))) = (𝐺𝑒𝑟𝑚𝑎𝑛𝑦) (𝐹𝑟𝑎𝑛𝑐𝑒, 𝐺𝑒𝑟𝑚𝑎𝑛𝑦) = 1 2 𝐴𝑅6(',)) is underestimated L
  • 9. Link prediction 2: GRank • Consider GPARs as entity ranking system • Define a scoring function whose arguments are Graph Pattern and a pair of entities: • When applying a GPARs to answer a query, create rankings of the candidate entities by their score • We can evaluate GPAR as an entity ranking system by evaluating output rankings.
  • 10. Link prediction 2: GRank • Distributed Ranking • Given a pattern GP and a query (h, r, ?), we obtain distributed rankings of entities according to their scores: 𝑑𝑟𝑎𝑛𝑘K,W = X 6 Y 0 When 𝑎 + 1 ≤ 𝑖 ≤ 𝑎 + 𝑏 Otherwise Entity j rank i 1 2 3 … a: スコアがentity j より⼤きいentityの数 b: スコアがentity j と同じentity の数
  • 11. Link prediction 2: GRank • To evaluate GPAR, define metric to evaluate distributed rankings of entities by generalizing the average precision • For a pattern GP and a training query (h, r, ?), the distributed precision at rank k is defined: Entity j rank i 1 2 … … rank k
  • 12. Link prediction 2: GRank • For a pattern GP and a training query (h, r, ?), the distributed average precision is defined: • The distributed mean average precision for a GPAR GP->r is defined: Entity j k 1 2 3 …
  • 13. Experiments: Baselines • Knowledge graph embedding models • Translation-based: TransE (Bordes et al. 2013), TrousE (Ebisu and Ichise, 2018) • Simple (h+r = t) and effective • Bilinear: RESCAL (Nickel et al., 2011) • Neural network-based: GCNs (Duvenaud et al., 2015) • Observed Feature models: predict based on observable features in the graph (???) • Node+LinkFeat (Toutanova et al., 2015) • only use on-hop information • PRA (Lao and Cohen, 2010) • use multi-hop information
  • 14. Results • WN18 from WordNet (1995): Well constructed. Few missing or wrong facts • FB15k from Freebase (2008): Many missing facts. • WN18 and FB15k have redundancy in the form of reverse relations -> WN18RR and FB15k-237: the inverse relation of other relations are removed • Graph pattern is restricted to connected and closed (No branch between x and y) • Limit the size of GP to be <= 3 Mean reciprocal rank Observed feature models The proportion of test queries whose corresponding entities are ranked in the top n of the obtained rankings Knowledge graph embedding models eliminating entities whose corresponding triples (except the target triple) were included in the training dataset.
  • 15. Results • Example of antecedent patterns for dMAP_tail which were given high ranks by GRank for FB15k high Low Low This only works when an individual has more than two siblings Individual’s parents are often missing in FB15k
  • 16. Conclusions • Defined GPAR for a knowledge graph. • Defined the standard confidence measures of GPARs for the link prediction. • Introduced GPro model • Introduced GRank model using distributed ranking -> SOTA • Future work: Extend GRank to use more complex pattern • Graph pattern size > 3