Relation extraction method based on explicit and implicit entity constraints
An implicit entity and relationship extraction technology, applied in character and pattern recognition, instrument, semantic tool creation, etc., can solve the problem of insufficient training and sparse data, alleviate the problem of gradient disappearance, improve the expansion ability, and improve the generalization ability. Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0034] See attached figure 1 , the present invention is based on the framework of the relationship extraction method of explicit and implicit entity constraints, and adopts the method of entity type attention mechanism to extract explicit entity constraint information. For each sentence in a package, entity constraint information that integrates sentence semantics can be extracted, and finally Through the sentence-level attention mechanism, the entity constraint information at the package level is obtained. The specific steps of relationship extraction are as follows:
[0035] Step 1: Data Preprocessing
[0036] Select a large-scale dataset marked by the heuristic of remote supervision, and combine the sentences aligned according to the same entity into a package, then segment each sentence in the package, and use the GloVe model to pre-train the word vector, each sentence will correspond to a matrix composed of word vectors, the specific operation is as follows:
[0037] 1-...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


