search

dice-group / dice-embeddings

Hardware-agnostic Framework for Large-scale Knowledge Graph Embeddings
MIT License
42 stars 12 forks source link

Combining Neural Link Predictor, Numerical Literal Predictor under LitCQD, #160

Open Demirrr opened 11 months ago

Demirrr commented 11 months ago

Inputs

  1. Train a neural link predictor (e.g. KECI)
  2. Numerical literal predictor (e.g. LiteralE or design a new one with us :) )

Combine (1) and (2) under LitCQD model/framework to tackle approximate query answering involving numerical literals.

More will be included

Demirrr commented 9 months ago

Training a model with byte_pair_encoding should lessen/mitigate this need. Needs to be validated