Semanlink - Knowledge Graph Embedding: A Survey of Approaches and Applications

Tags:

- "facts only" based embeddings
    - Translational Distance Models, distance-based scoring functions
       - TransE and co: entities and relations as points in vector space
           - TransH: relation-specific hyperplanes.
       - Gaussian Embeddings: entities and relations as random variables
    - Semantic matching models, similarity-based scoring functions
        - RESCAL (entities as a vector, relation as a matrix) and familiy
            - DistMult (relation as diagonal matrix)
            - Holographic Embeddings (HolE)
            - etc.
        - Matching with Neural Networks
            - SME, etc
- incorporating other info: entity types, relation paths,
textual descriptions, as well as logical rules
    - entity types
        - semantically smooth embedding
(SSE):  entities of the same type to stay close
to each other in the embedding space
        - type-embodied knowledge representation
learning (TKRL), which can handle hierarchical
entity categories and multiple category labels
- Applications in downstream tasks

Au sujet de ce document

sl:bookmarkOf : https://ieeexplore.ieee.org/document/8047276
sl:creationDate : 2019-05-27
sl:creationTime : 2019-05-27T18:42:06Z

Infos sur le fichier

Bookmark of: https://ieeexplore.ieee.org/document/8047276

Linked From

[2003.02320] Knowledge Graphs

Tags:

2020-03-07 A propos

Reasoning With Neural Tensor Networks for Knowledge Base Completion (2013)

Tags:

**Predicting the likely truth of additional facts based on existing facts in the knowledge base.**

> we introduce an expressive neural
tensor network suitable for reasoning over relationships between two entities.

Most similar work: [Bordes et al.](http://127.0.0.1:8080/semanlink/doc/2019/08/learning_structured_embeddings_) (2011)

Contributions:

1. new neural tensor
network (**NTN**) suitable for reasoning over relationships between two entities. Generalizes several previous neural network models and provides a more
powerful way to model relational information than a standard neural network layer.
2. a new way to represent entities in knowledge bases, as the
average of  their constituting word vectorss, allowing the sharing of statistical strength between the words describing
each entity (e.g., Bank of China and China).
3. incorporation of word vectors which are trained on large unlabeled text

> We **learn to modify word representations
via grounding in world knowledge**. This essentially allows us to analyze word embeddings and
query them for specific relations. Furthermore, the resulting vectors could be used in other tasks
such as named entity recognition or relation classification in natural language

**Makes use of entity name**: NTN
first learns word vectors from an auxiliary news corpus, and
then initializes the representation of each entity by averaging
the vectors of words contained in its name. For example,
**the embedding of AlfredHitchcock is initialized by the
average word vectors of “alfred” and “hitchcock”**... This kind of methods model textual information
separately from KG facts, and hence fail to leverage
interactions between them.[src](doc:2019/05/knowledge_graph_embedding_a_su)

2019-08-03 A propos

Documents with similar tags (experimental)

CIKM 2023 Tutorial--Reasoning beyond Triples: Recent Advances in Knowledge Graph Embeddings

Tags:

2024-04-01 A propos

[2308.00081] Towards Semantically Enriched Embeddings for Knowledge Graph Completion

Tags: