Automated Natural Language Processing (NLP) System
An Automated Natural Language Processing (NLP) System is a data processing system that solves an NLP task (by implementing an NLP algorithm).
- Context:
- It can range from being a Corpus-based NLP System to being a Heuristic NLP System.
- It can range from being a Semantic NLP System, to being a Syntactic NLP System, to being a Discourse-based NLP System, to being a Pattern-Matching NLP System.
- It can range from being a Rule-based NLP System, to being a Statistical NLP System to being a Neural NLP System,
- it can range from being a SVM-based NLP System, to being a HMM-based NLP System, to being a CRF-based NLP, to being a N-gram-based NLP System, to being a Neural NLP System.
- It can range from being an NLP-based Application to being an NLP Program.
- It can range from being an NLP Platform to being an NLP Toolkit to being an NLP Library.
- ...
- It can support a Text Mining System.
- It can be a part of an NLP Market.
- It can be based on an NLP Service.
- …
- Example(s):
- Domain-Specific NLP Systems, such as:
- a Legal NLP System for contract analysis and legal document processing.
- a Medical NLP System for clinical note interpretation and medical literature mining.
- a Financial NLP System for sentiment analysis of financial news and reports.
- a Scientific NLP System for automated literature review and hypothesis generation.
- a Technical Support NLP System for automating responses to customer queries.
- a Social Media NLP System for brand sentiment analysis and trend detection.
- an E-commerce NLP System for product categorization and review analysis.
- a Cybersecurity NLP System for threat detection in textual data.
- a Patent Analysis NLP System for innovation tracking and competitive intelligence.
- a Regulatory Compliance NLP System for policy adherence checking in corporate documents.
- Task-Specific NLP Systems, such as:
- a Morphological Analysis System.
- a Sentence Boundary Detection System.
- a Entity Mention Coreference Resolution System.
- a Lemmatization System, a Grammar Induction System.
- a Anaphora Reference Resolution System.
- a Tokenization System.
- a Discourse Integration System.
- a Part-Of-Speech Tagging System.
- a Text Chunking System.
- a Named Entity Recognition (NER) System.
- a Semantic Role Labeling (SRL) System.
- a Natural Laguage Parsing System.
- a Word Sense Disambiguation System.
- a NLP Annotation System.
- a Sentiment Analysis System.
- a Speech Recognition System.
- a Natural Language Understanding System.
- a Machine NL Translation System.
- a Natural Language Generation System, such as a dialoguing system.
- a Natural Language User Interface System.
- a Chatbot System, such as: ELIZA.
- ...
- a Method-Specific NLP System, such as:
- an LLM-based NLP System.
- a CRF-based NLP System.
- ...
- LUNAR System;
- BASEBALL System;
- DBPal System;
- AlLaDIn System;
- NaLIR System;
- PANTO System;
- Querix System;
- Aqualog System;
- …
- Domain-Specific NLP Systems, such as:
- Counter-Example(s):
- See: Controlled Natural Language, WikiText System, Large Language Model, Word Embedding Task, Transfer Learning in NLP, Explainable AI in NLP, Multilingual NLP Systems, Cross-lingual Transfer Learning.
References
2019
- (Wikipedia, 2019) ⇒ https://en.wikipedia.org/wiki/Natural_language_processing Retrieved:2019-2-17.
- Natural language processing (NLP) is a subfield of computer science, information engineering, and artificial intelligence concerned with the interactions between computers and human (natural) languages, in particular how to program computers to process and analyze large amounts of natural language data.
Challenges in natural language processing frequently involve speech recognition, natural language understanding, and natural language generation.
- Natural language processing (NLP) is a subfield of computer science, information engineering, and artificial intelligence concerned with the interactions between computers and human (natural) languages, in particular how to program computers to process and analyze large amounts of natural language data.
2011
- (Collobert et al., 2011b) ⇒ Ronan Collobert, Jason Weston, Léon Bottou, Michael Karlen, Koray Kavukcuoglu, and Pavel Kuksa. (2011). “Natural Language Processing (Almost) from Scratch.” In: The Journal of Machine Learning Research, 12.
- QUOTE: All the NLP tasks above can be seen as tasks assigning labels to words. The traditional NLP approach is: extract from the sentence a rich set of hand-designed features which are then fed to a standard classification algorithm, for example, a Support Vector Machine (SVM), often with a linear kernel. The choice of features is a completely empirical process, mainly based first on linguistic intuition, and then trial and error, and the feature selection is task dependent, implying additional research for each new NLP task. Complex tasks like SRL then require a large number of possibly complex features (e.g., extracted from a parse tree) which can impact the computational cost which might be important for large-scale applications or applications requiring real-time response.
Instead, we advocate a radically different approach: as input we will try to pre-process our features as little as possible and then use a multilayer neural network (NN) architecture, trained in an end-to-end fashion. The architecture takes the input sentence and learns several layers of feature extraction that process the inputs. The features computed by the deep layers of the network are automatically trained by backpropagation to be relevant to the task. We describe in this section a general multilayer architecture suitable for all our NLP tasks, which is generalizable to other NLP tasks as well.
Our architecture is summarized in Figure 1 and Figure 2. The first layer extracts features for each word. The second layer extracts features from a window of words or from the whole sentence, treating it as a sequence with local and global structure (i.e., it is not treated like a bag of words). The following layers are standard NN layers.
- QUOTE: All the NLP tasks above can be seen as tasks assigning labels to words. The traditional NLP approach is: extract from the sentence a rich set of hand-designed features which are then fed to a standard classification algorithm, for example, a Support Vector Machine (SVM), often with a linear kernel. The choice of features is a completely empirical process, mainly based first on linguistic intuition, and then trial and error, and the feature selection is task dependent, implying additional research for each new NLP task. Complex tasks like SRL then require a large number of possibly complex features (e.g., extracted from a parse tree) which can impact the computational cost which might be important for large-scale applications or applications requiring real-time response.
2008
- (Saranya, 2008) ⇒ S. K. Saranya. (2008). “Morphological Analyzer for Malayalam Verbs.” In: M. Tech Thesis, Amrita School of Engineering, Coimbatore.
- QUOTE: NLP problem can be divided into two tasks:
- Processing written text, using lexical, syntactic and semantic knowledge of the language as well as the required real world information.
- Processing spoken language, using all the information needed above plus additional knowledge about phonology as well as enough added information to handle the further ambiguities that arise in speech.(...)
- QUOTE: NLP problem can be divided into two tasks: