Matching syntactic-semantic graphs for semantic relation assignment

Năstase, Vivi; Śzpakowicz, Stan

doi:10.3115/1654758.1654775

Cited by 8 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Graph-theoretical objects can be treated as a natural way of modeling one's impressions of a text. As the graphical objects are able to preserve the connections or relationship between its constituents of varying granularity [15]. With the help of the two fundamental components of graph structures namely vertices and edges, we can represent a text, hardly losing any information.…”

Section: Graph-based Modeling For Natural Language Textsmentioning

confidence: 99%

See 1 more Smart Citation

A Graph Theoretical Preprocessing Step for Text Compression

Phukon¹,

Baruah²

2015

IJMUE

View full text Add to dashboard Cite

This paper presents CSGM 2 , a text preprocessing technique for compression purposes. It converts the original text into a word net (graph representation) and can retain the detailed contextual information such as word proximity. Specific directed graph is proposed to model this word net where words are stored in vertices and edges represent word transitions. The word net is fully capable of holding the natural word order in the original text and hence can be used directly for encoding purposes.

show abstract

Section: Graph-based Modeling For Natural Language Textsmentioning

confidence: 99%

“…These approaches represent the text by means of a graph, in which words or other text entities with meaningful relations are interconnected through vertices and edges. The information provided by the word order in a text is enough to construct a connected directed graph [15].…”

Section: Graph-based Modeling For Natural Language Textsmentioning

confidence: 99%

A Graph Theoretical Preprocessing Step for Text Compression

Phukon¹,

Baruah²

2015

IJMUE

View full text Add to dashboard Cite

show abstract

“…When trying to generate automatically favourable learning paths that match with the learner's needs the guidance should have a suitable balance between constraints for sustainability and freedom of association. Nastase and Szpakowicz (2006) introduced an incremental learning algorithm that effectively mimics the way in which a human reader accumulates knowledge and exploits it to process new text. The algorithm assigns a semantic relation to semantic units of text taken from a science book with guidance from a user and builds a simple syntactic-semantic graph surrounding the central concept and matching it with previously analysed text.…”

Section: Accumulating Knowledgementioning

confidence: 99%

Personalized Learning Paths Based on Wikipedia Article Statistics

Lahti¹

2010

Proceedings of the 2nd International Conference on Computer Supported Education

View full text Add to dashboard Cite

Abstract:We propose a new semi-automated method for generating personalized learning paths from the Wikipedia online encyclopedia by following inter-article hyperlink chains based on various rankings that are retrieved from the statistics of the articles. Alternative perspectives for learning topics are achieved when the next hyperlink to access is selected based on hierarchy of hyperlinks, repetition of hyperlink terms, article size, viewing rate, editing rate, or user-defined weighted mixture of them all. We have implemented the method in a prototype enabling the learner to build independently concept maps following her needs and consideration. A list of related concepts is shown in a desired type of ranking to label new nodes (titles of target articles for current hyperlinks) accompanied with parsed explanation phrases from the sentences surrounding each hyperlink to label directed arcs connecting nodes. In experiments the alternative ranking schemes well supported various learning needs suggesting new pedagogical networking practices. INTRODUCTIONTo enhance the quality and efficiency of automated information processing there is a need for new computational methods. The methods used for representing and modifying information with the computers should be made more compatible with the methods that are naturally used by humans to adopt and associate meanings. Thus developing illustrative adaptive computational methods that can be used in knowledge management in natural language and with intuitive visualizations should have a high priority in research agenda. We suggest that information available in structured online knowledge bases can serve as a valuable resource for computer-assisted learning in respect to the key concepts of curriculum and semantic relations linking them. We propose a new adaptive method to assist individual learning of core network of semantic relations in curriculum and later expanding this network further. We consider that still today the most valuable environment for learning is social collaboration in everyday life. We do not aim to challenge this traditional method but to complement it in a fruitful manner. We propose a new semi-automated method for generating personalized learning paths from the Wikipedia online encyclopedia by following inter-article hyperlink chains based on various rankings that are retrieved from the statistics of the articles. A learning path describes a structure of actions a learner has to perform in order to attain a competence or a competence profile (Janssen et al., 2008). In our proposal the learning paths are represented with concept maps. We are interested in methodology related to semantic navigation, intelligent tutoring systems and content-based filtering.Our method aims to ensure that an optimal coverage of concepts becomes provided to the learner. Especially our method aims to enable active explorations of conceptual relations taking into account various diverse perspectives that are based on different individual backgrounds and preference. Besides in...

show abstract

“…This need has ledto natural language processing and understanding. Numbers of terms are required to represent the capabilities of the method, they are as follow: Event resolution, redundancy reduction, labeling, knowledge base, word sense disambiguation, word sense induction semantic [8] relatedness and similarity measures [10].These term if achieve in its best in method for natural language processing we can get best and fast access to the information. The main goal of this paper is to represent the survey on the capabilities of the various graph based method [9] for natural language processing and natural language understanding.…”

Section: Introductionmentioning

confidence: 99%

Graph based Methods for Classification, Similarity and Representation using Noun, Verbs - A Survey

Patel¹,

Kyada²

2016

IJERT

View full text Add to dashboard Cite

This paper shows the survey of graph based methods for natural language processing [7]. Mainly the work focus on classification methods, similarity between the two phrases and representing this in terms of nouns verbs phrases. Classification done using clustering method, similarity is achieved by measuring similarity measures using various methods. Classification and measuring similarity basically depend on representation of the document which is based on nouns, verbs, phrases.

show abstract

Matching syntactic-semantic graphs for semantic relation assignment

Cited by 8 publications

References 15 publications

A Graph Theoretical Preprocessing Step for Text Compression

A Graph Theoretical Preprocessing Step for Text Compression

Personalized Learning Paths Based on Wikipedia Article Statistics

Graph based Methods for Classification, Similarity and Representation using Noun, Verbs - A Survey

Contact Info

Product

Resources

About