knowledge graph embedding for mining cultural heritage
play

Knowledge Graph Embedding for Mining Cultural Heritage Data Nada - PowerPoint PPT Presentation

Knowledge Graph Embedding for Mining Cultural Heritage Data Nada Mimouni and Jean-Claude Moissinac Telecom ParisTech Institut Mines Telecom January 24 th , 2019 DIG - LTCI Knowledge Graph Embedding for Mining Cultural Heritage Data 1 / 34


  1. Knowledge Graph Embedding for Mining Cultural Heritage Data Nada Mimouni and Jean-Claude Moissinac – Telecom ParisTech Institut Mines Telecom January 24 th , 2019 DIG - LTCI Knowledge Graph Embedding for Mining Cultural Heritage Data 1 / 34

  2. Project Data Method Experiments Conclusion Outline Project presentation 1 Data 2 Knowledge Graph Embedding 3 Entities extraction Context graph Graph walks and kernel Neural language model Using the model Experiments and preliminary results 4 Entity similarity and relatedness Entity matching Conclusion 5 Knowledge Graph Embedding for Mining Cultural Heritage Data 2 / 34

  3. Project Data Method Experiments Conclusion Outline Project presentation 1 Data 2 Knowledge Graph Embedding 3 Entities extraction Context graph Graph walks and kernel Neural language model Using the model Experiments and preliminary results 4 Entity similarity and relatedness Entity matching Conclusion 5 Knowledge Graph Embedding for Mining Cultural Heritage Data 3 / 34

  4. Project Data Method Experiments Conclusion Project presentation Knowledge Graph Embedding for Mining Cultural Heritage Data 4 / 34

  5. Project Data Method Experiments Conclusion Project presentation Knowledge Graph Embedding for Mining Cultural Heritage Data 5 / 34

  6. Project Data Method Experiments Conclusion Outline Project presentation 1 Data 2 Knowledge Graph Embedding 3 Entities extraction Context graph Graph walks and kernel Neural language model Using the model Experiments and preliminary results 4 Entity similarity and relatedness Entity matching Conclusion 5 Knowledge Graph Embedding for Mining Cultural Heritage Data 6 / 34

  7. Project Data Method Experiments Conclusion Data Gather data from institutions: Collect data respecting privacy Adopt homogeneous representations to make the data comparable Choose a model able to represent links between data Rely on external data: DataTourism , tourist office data on places and events OpenAgenda , and other event calendar Joconde database, and other cultural data General knowledge bases: DBPedia , Wikidata , ... Geographical knowledge bases: geonames , data on data.gouv.fr ... Knowledge Graph Embedding for Mining Cultural Heritage Data 7 / 34

  8. Project Data Method Experiments Conclusion A simple example of links generation Knowledge Graph Embedding for Mining Cultural Heritage Data 8 / 34

  9. Project Data Method Experiments Conclusion Objectives Questions: How to collect, integrate and enrich this complex and large amount of data? How to mine such type of data to extract useful information? Hypothesis: Integrate external data source to enhance the quality of the original data; Limit the analysis to a specified context help boosting performance. Knowledge Graph Embedding for Mining Cultural Heritage Data 9 / 34

  10. Project Data Method Experiments Conclusion Approach Represent instances as a set of n-dimensional numerical feature vectors Use representation with different ML tasks Adapt neural language model : Word2vec 1 Transform RDF graph into sequences of entities and relations 2 (sentences) Train the model and generate entity vectors 3 + Conserve the information in the original graph + Semantically similar/related entities have close vectors in the embedded space + Generate a reusable model, that could be enriched with new entities Knowledge Graph Embedding for Mining Cultural Heritage Data 10 / 34

  11. Project Data Method Experiments Conclusion Outline Project presentation 1 Data 2 Knowledge Graph Embedding 3 Entities extraction Context graph Graph walks and kernel Neural language model Using the model Experiments and preliminary results 4 Entity similarity and relatedness Entity matching Conclusion 5 Knowledge Graph Embedding for Mining Cultural Heritage Data 11 / 34

  12. Project Data Method Experiments Conclusion Knowledge graph embedding process Paris 1 CMN Musée KG Recommandation completion Input Data 2 Extract entities Similarity / Link Community 7 Relatedness prediction detection 3 Build context graph 4 Generate walks . . . . . . . . . . . . ... random tf-idf black-list kernel 6 V n V1 V 2 V 3 5 Train neural language model Entities feature vectors Knowledge Graph Embedding for Mining Cultural Heritage Data 12 / 34

  13. Project Data Method Experiments Conclusion Extract entities (2) Identify entities’ URIs from input data URI exist: read and identify URI from data files 1 URI ! exist: use entity name to build URI (dbpedia, frdbpedia, wikidata) 2 Knowledge Graph Embedding for Mining Cultural Heritage Data 13 / 34

  14. Project Data Method Experiments Conclusion Build context graph (3) For each entity URI: Build context from a generalized data source , ’around’ the entity Data source: e.g. DBpedia ’around’: get neighbours in the graph within α hops Consider the undirected graph α = 1 or 2 Define a black-list to ignore predicates and objects: very general, e.g. <http://www.w3.org/2002/07/owl#Thing> non-informative, e.g. <http://fr.dbpedia.org/resource/Mod` ele:P.> noisy, e.g. <http://www.w3.org/2000/01/rdf-schema#comment> Knowledge Graph Embedding for Mining Cultural Heritage Data 14 / 34

  15. Project Data Method Experiments Conclusion Merge context graphs (3) e 8 e 7 e 8 e 7 e 5 e 14 e 4 e 4 e 3 e 13 e 6 e y e x e 12 e 9 e 6 e 5 e 1 e 10 e 2 e 3 Context graph of entity e y Context graph of entity e x e 12 e 14 e 13 e 7 e 5 e y e 8 e 4 e 3 e 6 e x e 10 e 9 e 1 e 2 Global context graph Knowledge Graph Embedding for Mining Cultural Heritage Data 15 / 34

  16. Project Data Method Experiments Conclusion Generate walks (4) Paris 1 CMN Musée KG Recommandation completion Input Data Extract entities 2 Similarity / Link Community 7 Relatedness prediction detection 3 Build context graph 4 Generate walks . . . . . . . . . . . . ... tf-idf black-list kernel random 6 V n V1 V 2 V 3 5 Train neural language model Entities feature vectors Knowledge Graph Embedding for Mining Cultural Heritage Data 16 / 34

  17. Project Data Method Experiments Conclusion Random walk (4) Intuition: all neighbours are equally important for an entity Specify walk parameters nb-walks: number of walks (example: 500 walk) depth: number of hops in the graph (2, 4, 8) example: d=4 ⇒ e → p 1 → e 1 → p 2 → e 2 Specify the list of entities (all entities in the global context graph / a predefined list) For each entity: get a random list of direct neighbours 1 calculate the corresponding number of walks for each neighbour 2 recursively.. 3 Adjust the number of walks according to specific cases: if (nb-neighbours < nb-walks) : divide, get the entire part of the division, sum-up the rest and add it to a randomly selected neighbour if (nb-neighbours == 0) : transfer its nb-walks to another randomly selected neighbour Knowledge Graph Embedding for Mining Cultural Heritage Data 17 / 34

  18. Project Data Method Experiments Conclusion Tf-Idf graph walk (4) Intuition: Some neighbours are more important for an entity. Prioritize important neighbours by weighting their predicates. Calculate tf-idf weights for predicates tf : evaluate the importance of a predicate p for an entity e t o ( p , e ) = number of p occurrences for entity e t p ( e ) = number of predicates associated with e tf ( p , e ) = t o ( p , e ) / t p ( e ) idf : evaluate the importance of a predicate p on the whole graph D = number of entities in the graph d ( p ) = number of entities using predicate p idf ( p ) = log ( D / d ( p )) tfidf ( p , e ) = tf ( p , e ) ∗ idf ( p ) Knowledge Graph Embedding for Mining Cultural Heritage Data 18 / 34

  19. Project Data Method Experiments Conclusion Black-list walk (4) Intuition: some predicates are noisy (less important) for an entity Put weights on predicates: predicate in the black-list: weight = 0 (to ignore) other predicate: weight = 1 (to consider in the walk) Example: { http://dbpedia.org/ontology/wikiPageWikiLink } Knowledge Graph Embedding for Mining Cultural Heritage Data 19 / 34

  20. Project Data Method Experiments Conclusion Weisfeiler-Lehman kernel (4) Intuition: Weisfeiler-Lehman subtree RDF graph kernels capture (richer) information of an entire subtree in a single node. de Vries, Gerben K. D., ”A Fast Approximation of the Weisfeiler-Lehman Graph Kernel for RDF Data”, ECML PKDD 2013. Knowledge Graph Embedding for Mining Cultural Heritage Data 20 / 34

  21. Project Data Method Experiments Conclusion Weisfeiler-Lehman kernel (4) For each iteration, for each entity in the graph, get random walks of depth d After 1 iteration, graph G sequences: 1 − > 6 − > 11; 1 − > 6 − > 11 − > 13; 1 − > 6 − > 11 − > 10; ... 4 − > 11 − > 6; 4 − > 11 − > 13; 4 − > 11 − > 10; 4 − > 11 − > 10 − > 8; ... Ristoski, Paulheim, ”RDF2Vec: RDF Graph Embeddings for Data Mining”, ISWC 2016. Knowledge Graph Embedding for Mining Cultural Heritage Data 21 / 34

  22. Project Data Method Experiments Conclusion Neural language model (5,6) Word2vec A two-layer neural net that processes text Input: a text corpus (sentences) Output: a set of vectors (feature vectors for words in that corpus) Create neural embeddings for any group of discrete and co-occurring states → RDF data Knowledge Graph Embedding for Mining Cultural Heritage Data 22 / 34

Recommend


More recommend