Natural Language Processing
Natural Language Processing NLP is an interdisciplinary subject that involves artificial intelligence, linguistics, computer science and other disciplines. It explores the problem of allowing computers to process natural language.
Natural language processing is based on technologies and resources such as big data, knowledge graphs, machine learning, and linguistics to achieve the interaction between computers and natural language.
Based on the input and output of natural language, NLP can be divided into two technical fields:
- Computer input natural language corresponds toNatural Language Understanding ;
- Computer output natural language corresponds toNatural Language Generation .
Difficulties of NLP
The difficulties of NLP focus on ambiguity, robustness, knowledge dependence, context understanding, etc.
Currently, there are three methods to implement NLP: rule-based methods, statistical methods, and deep learning.
Main Applications of NLP
- Text to speech
- Speech synthesis
- Speech recognition
- Chinese word segmentation
- Part-of-speech tagging
- Parsing
- Natural language generation
- Text categorization
- Question answering
- Machine translation
- Automatic summarization
- Textual entailment
- Information retrieval
- Information extraction
- Text-proofing
NLP Development Trends
- Traditional syntactic-semantic rules have been questioned. With the rise of corpus construction and linguistics, large-scale real text processing has become the main direction of the NLP field.
- Statistical mathematics methods are gaining attention, and the field of NLP is increasingly using machine learning methods to acquire language knowledge;
- Emphasis is placed on both shallow and deep processing, and on both statistical and rule-based methods, thus forming a hybrid system;
- NLP places more and more emphasis on the application of vocabulary, and a strong tendency towards "lexicalism" has emerged. The construction of a vocabulary knowledge base has become a common concern.