RelEx System
A RelEx System is a Relation Extraction System that can generate dependency trees by extracting logical relationships within a English text document.
- AKA: RelEx Semantic Relation Extractor, RelEx Parser, RelEx Dependency Relationship Extractor.
- Context
- It is mainly core component is a Dependency Parser that is built on a Carnegie-Mellon Link Grammar Parser.
- It is a narrow-AI component of OpenCog.
- It can solve a RelEx Task by implementing a RelEx Algorithm.
- It can output framing/grounding relationships and provides various sorts of tagging.
- It is supported by a RelEx2Logic System to map the ouput data into an OpenCog Atom representation.
- It is supported by a RelEx2Frame System to map the output data into FrameNet Relationship Sets.
- It implements some of the ideas of Hudson's Word Grammar.
- It is an open-source project available at: https://github.com/opencog/relex
- It can allow the following output formats: Plain ASCII, RelEx Compact Format, RelEx OpenCog Format, OWL format.
- Example(s):
relation-extractor.sh
⇒ https://github.com/opencog/relex#relation-extractorsh- Input Sentence: The cat chased a snake.
singular(cat)
singular(snake)
_subj(chase, cat)
_obj(chase, snake)
past(chase)
- Counter-Example(s):
- See: Natural Language Processing System, Parsing System, Semantic Relationship, Natural Language Comprehension System, Semantic Analysis System.
References
2019a
- (OpenCog, 2019) ⇒ https://wiki.opencog.org/w/RelEx_Dependency_Relationship_Extractor Retrieved: 2019-10-13.
- QUOTE: RelEx, a narrow-AI component of OpenCog, is an English-language semantic dependency relationship extractor, built on the Carnegie-Mellon Link Grammar parser. It uses a series of graph rewriting rules to identify subject, object, indirect object and many other syntactic dependency relationships between words in a sentence. That is, it generates the dependency trees of a dependency grammar. The set of dependency relations it employs are quite similar to those of Dekang Lin's MiniPar, the Stanford parser and Google's SyntaxNet. (It even has an explicit compatibility mode with the Stanford parser). It is inspired in part by the ideas of Hudson's Word Grammar.
Unlike other dependency parsers, RelEx attempts a greater degree of semantic normalization: for questions, comparatives, entities, and for modifying clauses, and sub-ordinate clauses, whereas other parsers (such as the Stanford parser) stick to a literal presentation of the syntactic structure of text. For example, RelEx pays special attention to determining when a sentence is hypothetical or speculative, and to isolating the query variables from a question. Both of these aspects are intended to make RelEx well-suited for question-answering and semantic comprehension/reasoning systems. In addition, RelEx makes use of feature tagging, to tag words with part-of-speech, noun-number, verb-tense, gender, etc. As of this writing, RelEx parses text nearly four times faster than the Stanford parser; and it now provides a “compatibility mode", wherein it can generate the same relations as the Stanford parser.
- QUOTE: RelEx, a narrow-AI component of OpenCog, is an English-language semantic dependency relationship extractor, built on the Carnegie-Mellon Link Grammar parser. It uses a series of graph rewriting rules to identify subject, object, indirect object and many other syntactic dependency relationships between words in a sentence. That is, it generates the dependency trees of a dependency grammar. The set of dependency relations it employs are quite similar to those of Dekang Lin's MiniPar, the Stanford parser and Google's SyntaxNet. (It even has an explicit compatibility mode with the Stanford parser). It is inspired in part by the ideas of Hudson's Word Grammar.
2019b
- (OpenCog, 2019) ⇒ https://wiki.opencog.org/w/Dependency_relations Retrieved: 2019-10-13.
- QUOTE: As a general idea, RelEx attempts a greater degree of semantic abstraction; that is, it is less aimed at presenting the syntactic structure of sentence, than it is in presenting its semantic content. This shows itself in several ways:
- RelEx attempts basic entity extraction, and thus avoids generating nn noun modifier relations for named entities.
- RelEx will collapse the object and complement of a preposition into one. Stanford will do this for some, but not all relationships.
- RelEx will convert passive subjects into objects, and instead indicate passiveness by tagging the verb with a passive tense feature.
- RelEx avoids generating copulas, if at all possible, and instead indicates copular relations as predicative adjectives, or in other ways.
- RelEx extracts semantic variables from questions, with the intent of simplifying question answering. For example, Where is the ball? generates _pobj(_%atLocation, _$qVar) _psubj(_%atLocation, ball), which can then pattern-match a plausible answer: _pobj(under, couch).
- RelEx attempts to extract comparison variables.
- QUOTE: As a general idea, RelEx attempts a greater degree of semantic abstraction; that is, it is less aimed at presenting the syntactic structure of sentence, than it is in presenting its semantic content. This shows itself in several ways:
2016
- (Github, 2016) ⇒ https://github.com/opencog/relex#relex-semantic-relation-extractor
- QUOTE: RelEx is a dependency parser for the English language. It extracts dependency relations from Link Grammar, and adds some shallow semantic analysis. The primary use of RelEx is as a language input front-end to the OpenCog artificial general intelligence system.
There are multiple inter-related parts to RelEx. The core component extracts the dependency relationships. An experimental module provides some simple anaphora resolution suggestions. Output is provided in various formats, including one format suitable for later batch post-processing, another format suitable for input to OpenCog, and an W3C OWL format. There are also a small assortment of perl scripts for cleaning up web and wiki pages, &c.
- QUOTE: RelEx is a dependency parser for the English language. It extracts dependency relations from Link Grammar, and adds some shallow semantic analysis. The primary use of RelEx is as a language input front-end to the OpenCog artificial general intelligence system.
2014
- (Vepstas & Goertzel, 2014) ⇒ Linas Vepstas, and Ben Goertzel. (2014). "Learning Language from a Large (Unannotated) Corpus". eprint arXiv:1401.3372
- QUOTE: Just as our approach to syntax learning is heavily influenced by our work with Link Grammar. our approach to semantics is heavily influenced by our work on the RelEx system [RVG05, LGE10, GPPG06, LGK+12], which maps the output of the Link Grammar parser into a more abstract semantic form. Prototype systems [GPA+10, LGK+12] have also been written mapping the output of RelEx into even more abstract semantic form, consistent with the semantics of the Probabilistic Logic Networks [GIGH08] formalism as implemented in the OpenCog [HG08] framework. These systems are largely based on hand-coded rules, and thus not in the spirit of language learning pursued in this proposal. However, they display the same structure that we assume here; the difference being that here we specify a mechanism for learning the linguistic content that fills in the structure via unsupervised corpus learning, obviating the need for hand-coding.
2013
- (Silva et al., 2013) ⇒ Nayanthara De Silva, Chiran N. Fernando, M. K. D. T. Maldeniya, Dileep Wijeratne, A. S. Perera, and Ben Goertzel. (2013). “Mapping Dependency Relationships Into Semantic Frame Relationships.”
- QUOTE: RelEx, which is a component, developed for the OpenCog [1] framework is an English-language semantic dependency relationship extractor, built on the Carnegie-Mellon Link Grammar parser [2]. Subject, object, indirect object and many other syntactic dependency relationships between words in a sentence can be identified by RelEx.
2012
- (Lian et al., 2012) ⇒ Ruiting Lian, Ben Goertzel, Shujing Ke, Jade O'Neill, Keyvan Sadeghi, Simon Shiu, Dingjie Wang, Oliver Watkins, and Gino Yu. (2012). “Syntax-Semantic Mapping for General Intelligence: Language Comprehension As Hypergraph Homomorphism, Language Generation As Constraint Satisfaction.” In: Proceedings of the 5th International Conference on Artificial General Intelligence. ISBN:978-3-642-35505-9 doi:10.1007/978-3-642-35506-6_17
- QUOTE: RelEx is an English-language semantic relationship extractor, designed to postprocess the output of the link parser. It can identify subject, object, indirect object and many other dependency relationships between words in a sentence; it generates dependency trees, resembling those of dependency grammars (...)
RelEx currently works via creating a tree with a
FeatureNode
corresponding to each word in the sentence, and then applying a series of rules to update the entries in thisFeatureNode
. The rules transform combinations of link parser links into RelEx dependency relations, sometimes acting indirectly via dynamics wherein one rule changes a feature in a word’sFeatureNode
, and another rule then takes an action based on the changes the former rule made.
- QUOTE: RelEx is an English-language semantic relationship extractor, designed to postprocess the output of the link parser. It can identify subject, object, indirect object and many other dependency relationships between words in a sentence; it generates dependency trees, resembling those of dependency grammars (...)
2010a
- (Cer et al., 2010) ⇒ Daniel Cer, Marie-Catherine de Marneffe, Daniel Jurafsky, and Christopher D. Manning l. (2010). “Parsing to Stanford Dependencies: Trade-offs Between Speed and Accuracy.” In: Proceedings of the Seventh International Conference on Language Resources and Evaluation (LREC'10).
- QUOTE: We examine algorithms specifically designed for dependency parsing (Nivre, Nivre Eager, Covington, Eisner, and RelEx) as well as dependencies extracted from constituent parse trees created by phrase structure parsers (Charniak, Charniak-Johnson, Bikel, Berkeley and Stanford).
2010b
- (Vepstas et al., 2010) ⇒ Linas Vepstas, Mike Ross, and Ben Goertzel (2010). "Identifying Speculation from Syntax". eprint published online on 2010-05-23.
- QUOTE: The RelEx semantic relation extractor is a dependency parser that generates feature markup identifying hypothetical and speculative statements, questions and clauses. These are identified by means of fixed, linguist-generated rules applied to syntactically parsed sentences. By using purely syntactic information, obtained from single sentences, this approach provides a baseline for extracting the semantic content of speculative sentences (...)
In the sentence "The dog will eat it", the verb "eat" gets marked as being hypothetical – the dog hasn’t eaten it yet, and there are no assurances that it will happen – only that it could happen. In this case, an imperative or future tense verb is sufficient to indicate the speculative nature of the statement. RelEx designates hypothetical or speculative statements by means of a feature tag – a single bit of markup,
HYP
, that is used to tag the verb making the hypothetical assertion. The generated output is thus:_subj(eat, dog)
_obj(eat, it)
tense(eat, imperative)
HYP(eat, T)
Here,
_subj
and_obj
are standard subject and object dependencies; as usual, the head word is always listed first. Thetense
andHYP
are feature tags: they identify a property, naming the word as the first argument, and the property value as second. Here,T
, for ’true’, simply means that the property is present (...)
- QUOTE: The RelEx semantic relation extractor is a dependency parser that generates feature markup identifying hypothetical and speculative statements, questions and clauses. These are identified by means of fixed, linguist-generated rules applied to syntactically parsed sentences. By using purely syntactic information, obtained from single sentences, this approach provides a baseline for extracting the semantic content of speculative sentences (...)
2008a
- (Hart & Goertzel, 2008) ⇒ David Hart, and Ben Goertzel. (2008). “OpenCog: A Software Framework for Integrative Artificial General Intelligence.” In: Proceedings of the 2008 conference on Artificial General Intelligence 2008: Proceedings of the First AGI Conference. ISBN:978-1-58603-833-5
- QUOTE: RelEx integration, a natural language comprehension system that intakes English sentences and outputs logical relationships suitable for analysis via AI algorithms. RelEx is to be released soon in conjunction with the Wikia Search project.
2008b
- (Goertzel & Garis, 2008) ⇒ Ben Goertzel, and Hugo de Garis. (2008). “XIA-MAN: An Extensible, Integrative Architecture for Intelligent Humanoid Robotics.”
- QUOTE:, in its comprehension aspect, takes English text and maps that text into abstract logical relations, in the Atom format utilized internally by the NCE and OpenCog. Generally speaking it produces multiple interpretations (logical relation sets) for each sentence it processes, and the task of selecting the contextually appropriate interpretation is left to the cognition engine itself. Also, the cognition engine is relied upon to correct errors RelEx may make in areas such as word sense disambiguation and reference resolution. It is anticipated that the sensory data gathered by a robot, regarding the physical and social context of instances of linguistic usages it produces or hears, may provide data helpful to the cognition engine in executing the linguistic tasks of interpretation-selection, reference resolution and sense disambiguation.
Next, in its generation aspect, RelEx maps logical relation sets (Atom sets) into sets of English sentences (...)
- QUOTE:, in its comprehension aspect, takes English text and maps that text into abstract logical relations, in the Atom format utilized internally by the NCE and OpenCog. Generally speaking it produces multiple interpretations (logical relation sets) for each sentence it processes, and the task of selecting the contextually appropriate interpretation is left to the cognition engine itself. Also, the cognition engine is relied upon to correct errors RelEx may make in areas such as word sense disambiguation and reference resolution. It is anticipated that the sensory data gathered by a robot, regarding the physical and social context of instances of linguistic usages it produces or hears, may provide data helpful to the cognition engine in executing the linguistic tasks of interpretation-selection, reference resolution and sense disambiguation.
2007
- (Fundel et al., 2007) ⇒ Katrin Fundel, Robert Küffner, and Ralf Zimmer. (2007). “RelEx- Relation Extraction Using Dependency Parse Trees.” In: Bioinformatics Journal, 23(3). doi:10.1093/bioinformatics/btl616
- QUOTE: The RelEx work-flow (Figure 1) extracts directed qualified relations starting from free-text sentences. RelEx requires a synonym dictionary (Fundel and Zimmer, 2006) containing gene and protein names, and a list of restriction-terms1 that are used to describe relations of interest.
Fig. 1. The work-flow of RelEx is subdivided into preprocessing, relation extraction and relation filtering leading from the original free-text sentences to directed, qualified relations. Preprocessing is based on publicly available tools and named entity identification. Candidate relations are extracted according to rules applied on chunk dependency trees and original sentences, and subjected to filtering steps.
- QUOTE: The RelEx work-flow (Figure 1) extracts directed qualified relations starting from free-text sentences. RelEx requires a synonym dictionary (Fundel and Zimmer, 2006) containing gene and protein names, and a list of restriction-terms1 that are used to describe relations of interest.