Sentence analysis tools are used to identify syntactic elements in text and then automatically construct pattern rules from these elements. To extract patterns with better coverage and accuracy, larger corpora, more formats of patterns, and more efficient ways of extraction have been explored by later work in pattern extraction. Automatically extracted patterns may have mistakes so these methods require examination from human experts. 

Pattern extraction methods


SREs reduce human efforts and bring better coverage than pattern rules. 
The following are types of SREs:
- Feature-based methods
- Kernel-based methods
- Graphical methods
- Embedding models

Statistical Relation Extraction Models 

Most RE methods work in a simplified setting, and mainly focus on training models with large amounts of human annotations to classify two given entities within one sentence into pre-defined relations. To build an effective and robust RE system that can be deployed in the real world, more complex scenarios must be further investigated.  
- Collecting high-quality human annotation is expensive and time-consuming
- Many long-tail relations cannot provide large amounts of training examples 
- Most facts are expressed by long context consisting of multiple sentences
- Using a predefined set to cover those relations with open-ended growth is not easy

Drawbacks of existing RE methods


NRE models introduce neural networks to automatically extract semantic features from text, and can effectively capture textual information and generalize to a wider range of data. NRE studies mainly focus on designing and using different network architectures to capture relational semantics within text. NRE differs from SREs by using word embeddings and position embeddings as inputs so that semantic meanings of words can be combined with a specification of the relative distances between words and entities. 


Deep Learning - Neural relation extraction models (NRE)

The following approaches have been explored for relation extraction, and these methods have bridged the gap between unstructured text and structured knowledge. 
- Pattern mining
- Statistical approaches 
- Deep learning

Columbia University

Relation extraction is an important task in IE. Relation extraction is a task which aims at extracting relational facts from plain text by classifying relations between entity mentions in text from their context after identifying them. Relation extraction can benefit down-stream applications such as knowledge graph completion, search engine, and questions answering.

Relation extraction (RE)

Han, X., Gao, T., Lin, Y., Peng, H., Yang, Y., Xiao, C., … Zhou, J. (2020). More Data, More Relations, More Context and More Openness: A Review and Outlook for Relation Extraction. doi:10.48550/ARXIV.2004.03186


More Data, More Relations, More Context and More Openness: A Review and Outlook for Relation Extraction

Relational facts structure knowledge in the format of triplets. These facts represent an important role of human knowledge and are explicit or implicit in text. An example of an explicit representation is the sentence “Bill Gates co-founded Microsoft” which indicates the fact (Microsoft Inc., founded by, Bill Gates). An example of an implicit representation is that (USA, contains, New York) can be inferred from “Hamilton made its debut in New York, USA”. 

Relational facts

A typical RE setting is given a sentence marked by two entities, and models need to classify the sentence into one of the predefined relations. A complete relation extraction system consists of: 
- named entity recognizer to identify named entities  
- Entity linker to link entities to existing knowledge graphs 
- Relational classifier to determine relations between entities by given context
Identifying the relation is the most crucial and most difficult task because it requires models to understand the semantics of the context. Therefore, RE generally focuses on researching the relational classification side of the RE systems. 

Learn Before

Related

Learn After