Learning to Generate CGs from Domain Specific Sentences

Zhang, Lei; Yu, Yong

doi:10.1007/3-540-44583-8_4

Cited by 21 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have incorporated additional financial terms in the parser's dictionary to cater for the special needs arising in the problem domain as discussed in section 1 of this paper. We propose to use the LGP because; there exist a structure similarity to conceptual graphs hence it is easier to map the obtained structure to conceptual graphs [12].…”

Section: Parsingmentioning

confidence: 99%

Conceptual graph formalism for financial text representation

Kamaruddin

Bakar

Hamdan

et al. 2008

2008 International Symposium on Information Technology

View full text Add to dashboard Cite

We present an approach to automatically transform a financial text into conceptual graph formalism. The approach exploits the constituent structure of sentences and general English grammar rules to perform the transformation. We suggest face validation and traces as the evaluation method to be performed on the resulting formalism to validate its accuracy. We also discuss the potential manipulation and application of the constructed conceptual graph database.

show abstract

Section: Parsingmentioning

confidence: 99%

Conceptual graph formalism for financial text representation

Kamaruddin

Bakar

Hamdan

et al. 2008

2008 International Symposium on Information Technology

View full text Add to dashboard Cite

show abstract

“…In this sense, cgs 50 provide a tool to make commonalities explicit and to derive knowledge, inferring sub and super-concept relationships. The application of the model to km has become of increasing interest in the last decade 8,29,58 , although practical proposals have been limited to restricted domains, looking for both grammatical simplicity and a well-known vocabulary in texts: patent claims 43 , financial statements 34 or computer science 38,47 . The initial cause of these limitations is the difficulty to automatically generate cgs from raw texts, entailing a manual processing of the documents.…”

Section: The State-of-the-artmentioning

confidence: 99%

Intelligent Retrieval for Biodiversity

Ferro

Fernández-Gavilanes

González

et al. 2016

Int. J. Artif. Intell. Tools

View full text Add to dashboard Cite

A proposal for intelligent retrieval in the biodiversity domain is described. It applies natural language processing to integrate linguistic and domain knowledge in a mathematical model for information management, formalizing the notion of semantic similarity in different degrees. The goal is to provide computational tools to identify, extract and relate not only data but also scientific notions, even if the information available to start the process is not complete. The use of conceptual graphs as a basis for interpretation makes it possible to avoid the use of classic ontologies, whose start-up requires costly generation and maintenance protocols and also unnecessarily overload the accessing task for inexpert users. We exploit the automatic generation of these structures from raw texts through graphical and natural language interaction, at the same time providing a solid logical and linguistic foundation to sustain the curation of databases.

show abstract

“…LGP is used because; there exist a structure similarity to conceptual graphs; hence it is easier to map the obtained structure to conceptual graphs [67]. Suchanek et al [68] report that the LGP provides a much deeper semantic structure than the standard context-free parsers.…”

Section: Transforming Extracted Sentence Into Cgifmentioning

confidence: 99%

Deviation detection in text using conceptual graph interchange format and error tolerance dissimilarity function

Kamaruddin

Hamdan

Bakar

et al. 2012

IDA

View full text Add to dashboard Cite

The rapid increase in the amount of textual data has brought forward a growing research interest towards mining text to detect deviations. Specialized methods for specific domains have emerged to satisfy various needs in discovering rare patterns in text. This paper focuses on a graph-based approach for text representation and presents a novel error tolerance dissimilarity algorithm for deviation detection. We resolve two non-trivial problems, i.e. semantic representation of text and the complexity of graph matching. We employ conceptual graphs interchange format (CGIF)-a knowledge representation formalism to capture the structure and semantics of sentences. We propose a novel error tolerance dissimilarity algorithm to detect deviations in the CGIFs. We evaluate our method in the context of analyzing real world financial statements for identifying deviating performance indicators. We show that our method performs better when compared with two related text based graph similarity measuring methods. Our proposed method has managed to identify deviating sentences and it strongly correlates with expert judgments. Furthermore, it offers error tolerance matching of CGIFs and retains a linear complexity with the increasing number of CGIFs.

show abstract

Learning to Generate CGs from Domain Specific Sentences

Cited by 21 publications

References 7 publications

Conceptual graph formalism for financial text representation

Conceptual graph formalism for financial text representation

Intelligent Retrieval for Biodiversity

Deviation detection in text using conceptual graph interchange format and error tolerance dissimilarity function

Contact Info

Product

Resources

About