SlideShare une entreprise Scribd logo
1  sur  20
Télécharger pour lire hors ligne
Università degli studi di Bari “Aldo Moro”
                              Dipartimento di Informatica




             ConNeKTion: A Tool for Exploiting Conceptual
               Graphs Automatically Learned from Text
                                F. Leuzzi, S. Ferilli, F. Rotella
L.A.C.A.M.                 {fabio.leuzzi, stefano.ferilli, fulvio.rotella}@uniba.it
http://lacam.di.uniba.it

                           9th Italian Research Conference on Digital Libraries
                                    Università la Sapienza - Rome, Italy
                                       January 31 - February 1, 2013
Overview

             ●   Introduction & Objectives
             ●   Tool overview
             ●   Knowledge Representation Formalism
             ●   Relevant concepts
             ●   Information Retrieval
             ●   Reasoning by Association
             ●   Exploiting Tool
             ●   Conclusions & Future Works



ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     2
Introduction
      Some repositories leave the responsibility of quality to the authors.


                                                  +
                   Anybody can produce and distribute documents.


                                                  =
              Possible low average quality of the repository contents.




The study, understanding and exploitation of the content of a digital library,
with the aim to easily explore the semantic content of huge amounts of text.



ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     3
Introduction
         Possible solution:
 ●       Natural Language Processing systems
     ●    Provide the grammatical structures contained in text
 ●       Knowledge Representation formalisms
     ●    Semantic networks
 ●       Graph learning techniques
     ●    To obtain a semantic network starting from the text
 ●       In order to satisfy the information needs, the knowledge base
         can be exploited:
     ●    To make summarizations
     ●    To reason with it
     ●    ...

ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     4
Objectives

                                 Improving fruition of a DL


         ●       Use of a tool providing advanced functionalities
             ●       Mixed strategy for relevant concept recognition
             ●       Semantic approach to information retrieval
             ●       Automatic inference over the acquired knowledge
                 ●    Reasoning by association




ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     5
Tool overview




ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     6
Knowledge representation
                       formalism
    Only subject, verb and complement have been considered.
    ●   Subjects and complements → concepts
    ●   Verbs → relations between them




               subject,                                    subject,
                verb,                                    complement
             complement



    The frequency of arcs between the concepts in positive and negative
    sentences has been taken into account.
    ●   Enrich the representation formalism
    ●   Give robustness to our solution through a statistical approach
ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     7
Relevant Concepts

   ● Relevant nodes are sought in the graph

   ● Mixed strategy
      ●       Semantic network structure
      ●       EM clustering provided by Weka
      ●       Keyword Extraction
          ●    Quantitative approach based on co-occurrences
          ●    Qualitative approach exploiting WordNet
          ●    Psychological approach based on principles of an effective
               presentation
   ● Components empirically weighted



ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     8
Information Retrieval
   ●       Word Sense Disambiguation
       ●       One Domain per Discourse assumption: many uses of a word in a
               coherent portion of text tend to share the same domain
       ●       Prevalent domain individuation
           ●       Extraction of all synsets for each term
           ●       Extraction of all domains for each synset
           ●       Choice of prevalent domain synset
   ●       Pairwise Complete Link Agglomerative Clustering
       ●       Each synset generates a singleton cluster
       ●       For each pair of clusters
           ●       If the complete link property holds
               ●   Merge the involved clusters



ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     9
Information Retrieval
   ●       Multi-strategy Similarity Measure on WordNet
   ●       3 components summed and normalized in ]0,1[
       ●       depth (ancestors)
       ●       breadth (direct neighbors)
       ●       breadth (inverse neighbors)
   ●       Document Partitioning
       ●       For each document
           ●    Each synset votes for a cluster
   ●       User Query Processing
       ●       Brute force WSD to find the best synsets combination
       ●       Best combination used to return a ranked list of clusters
       ●       Each cluster has a list of related documents obtained by the Document
               Partitioning phase


ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                      10
Reasoning ‘by association’
                     Breadth-First Search
         Given two nodes (concepts), a Breadth-First Search starts from
         both nodes, the former searches the latter's frontier and vice
         versa, until the two frontiers meet. Then the path is restored
         going backward to the roots in both directions.




         We also provide the number of positive/negative instances, and
         the corresponding ratios over the total to help understanding
         different gradations (permitted, prohibited, typical, rare, etc.) of
         actions between two objects.

ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     11
Reasoning ‘by association’
                     Breadth-First Search
   The table below shows a sample of possible outcomes.
   E.g., an interpretation of case 1 can be:

   “the young looks television that talks about (and criticizes) facebook,
   because it typically does not help (rather distracts) schoolwork”.




ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     12
Reasoning ‘by association’
                   Probabilistic approach
  Real world data are typically noisy
  and uncertain → need for strategies
  that soften the classical rigid logical
  reasoning



  Defined a formalism based on ProbLog language: pi :: fi
    ●   fi : ground literal of the form link (subject, verb, complement)
    ●   pi : ratio between the
        sum of all examples for
        which fi holds and the sum
        of all possible links between
        subject and complement

ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     13
ConNeKTion




ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     14
ConNeKTion




ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     15
ConNeKTion




ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     16
ConNeKTion




ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     17
ConNeKTion




ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     18
Conclusions

  ConNeKTion allows to learn conceptual graphs from plain text and to
  enrich them by finding concept generalizations.


  The resulting graph can be used for several purposes:
  ●   finding relationships between concepts (if any)
  ●   filtering the concepts from a particular perspective
  ●   relevant concepts recognition and information retrieval


  A suitable control panel is provided for the user to comfortably carry out
  these activities.



ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                     19
Future Works
  We plan to improve the natural language text pre-processing using anaphora
  resolution in order to replace, where possible, pronouns with the explicit concept
  they express.


  All functionalities have parameters set empirically. A criteria for automatical
  setting of suitable parameters is needed.


  The preseted functionalities are based on the exploitation of WordNet. A strategy
  to make the operators WordNet free can be desirable.


  We also wish to extend the reasoning operators by adding an argumentation
  operator, that could exploit probabilistic weights, intended as a rate of reliability,
  to provide support or attack to a given statement.


ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text
F. Leuzzi, S. Ferilli, F. Rotella                                                      20

Contenu connexe

Tendances

Rethinking Critical Editions of Fragments by Ontologies
Rethinking Critical Editions of Fragments by OntologiesRethinking Critical Editions of Fragments by Ontologies
Rethinking Critical Editions of Fragments by Ontologies
Matteo Romanello
 
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from TextCooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
University of Bari (Italy)
 

Tendances (11)

Cooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from TextCooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
 
textTOvec: Deep Contextualized Neural Autoregressive Topic Models of Language...
textTOvec: Deep Contextualized Neural Autoregressive Topic Models of Language...textTOvec: Deep Contextualized Neural Autoregressive Topic Models of Language...
textTOvec: Deep Contextualized Neural Autoregressive Topic Models of Language...
 
Pankaj Gupta CV / Resume
Pankaj Gupta CV / ResumePankaj Gupta CV / Resume
Pankaj Gupta CV / Resume
 
Rethinking Critical Editions of Fragments by Ontologies
Rethinking Critical Editions of Fragments by OntologiesRethinking Critical Editions of Fragments by Ontologies
Rethinking Critical Editions of Fragments by Ontologies
 
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from TextCooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
 
An Approach to Automated Learning of Conceptual Graphs from Text
An Approach to Automated Learning of Conceptual Graphs from TextAn Approach to Automated Learning of Conceptual Graphs from Text
An Approach to Automated Learning of Conceptual Graphs from Text
 
The role of linguistic information for shallow language processing
The role of linguistic information for shallow language processingThe role of linguistic information for shallow language processing
The role of linguistic information for shallow language processing
 
The Semantic Web #8 - Ontology
The Semantic Web #8 - OntologyThe Semantic Web #8 - Ontology
The Semantic Web #8 - Ontology
 
Centroid-based Text Summarization through Compositionality of Word Embeddings
Centroid-based Text Summarization through Compositionality of Word EmbeddingsCentroid-based Text Summarization through Compositionality of Word Embeddings
Centroid-based Text Summarization through Compositionality of Word Embeddings
 
The Semantic Web #7 - RDF Semantics
The Semantic Web #7 - RDF SemanticsThe Semantic Web #7 - RDF Semantics
The Semantic Web #7 - RDF Semantics
 
How to Ground A Language for Legal Discourse In a Prototypical Perceptual Sem...
How to Ground A Language for Legal Discourse In a Prototypical Perceptual Sem...How to Ground A Language for Legal Discourse In a Prototypical Perceptual Sem...
How to Ground A Language for Legal Discourse In a Prototypical Perceptual Sem...
 

En vedette

En vedette (8)

A Run Length Smoothing-Based Algorithm for Non-Manhattan Document Segmentation
A Run Length Smoothing-Based Algorithm for Non-Manhattan Document SegmentationA Run Length Smoothing-Based Algorithm for Non-Manhattan Document Segmentation
A Run Length Smoothing-Based Algorithm for Non-Manhattan Document Segmentation
 
Recognising the Social Attitude in Natural Interaction with Pedagogical Agents
Recognising the Social Attitude in Natural Interaction with Pedagogical AgentsRecognising the Social Attitude in Natural Interaction with Pedagogical Agents
Recognising the Social Attitude in Natural Interaction with Pedagogical Agents
 
Recognising the Social Attitude in Natural Interaction with Pedagogical Agents
Recognising the Social Attitude in Natural Interaction with Pedagogical AgentsRecognising the Social Attitude in Natural Interaction with Pedagogical Agents
Recognising the Social Attitude in Natural Interaction with Pedagogical Agents
 
A Domain Based Approach to Information Retrieval in Digital Libraries - Rotel...
A Domain Based Approach to Information Retrieval in Digital Libraries - Rotel...A Domain Based Approach to Information Retrieval in Digital Libraries - Rotel...
A Domain Based Approach to Information Retrieval in Digital Libraries - Rotel...
 
Take your sbdc online
Take your sbdc onlineTake your sbdc online
Take your sbdc online
 
ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned fro...
ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned fro...ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned fro...
ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned fro...
 
A Run Length Smoothing-Based Algorithm for Non-Manhattan Document Segmentation
A Run Length Smoothing-Based Algorithm for Non-Manhattan Document SegmentationA Run Length Smoothing-Based Algorithm for Non-Manhattan Document Segmentation
A Run Length Smoothing-Based Algorithm for Non-Manhattan Document Segmentation
 
A Domain Based Approach to Information Retrieval in Digital Libraries - Rotel...
A Domain Based Approach to Information Retrieval in Digital Libraries - Rotel...A Domain Based Approach to Information Retrieval in Digital Libraries - Rotel...
A Domain Based Approach to Information Retrieval in Digital Libraries - Rotel...
 

Similaire à ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text

Cooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from TextCooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
University of Bari (Italy)
 
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from TextCooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
Fulvio Rotella
 
Timo Honkela: Multimodally Grounded Translation by Humans and Machines
Timo Honkela: Multimodally Grounded Translation by Humans and MachinesTimo Honkela: Multimodally Grounded Translation by Humans and Machines
Timo Honkela: Multimodally Grounded Translation by Humans and Machines
Timo Honkela
 
Listening comprehension in efl teaching
Listening comprehension in efl teachingListening comprehension in efl teaching
Listening comprehension in efl teaching
mora-deyanira
 
Listening Comprehension in EFL Teaching
Listening Comprehension in EFL TeachingListening Comprehension in EFL Teaching
Listening Comprehension in EFL Teaching
mora-deyanira
 

Similaire à ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text (20)

Timo Honkela: From Patterns of Movement to Subjectivity of Understanding
Timo Honkela: From Patterns of Movement to Subjectivity of UnderstandingTimo Honkela: From Patterns of Movement to Subjectivity of Understanding
Timo Honkela: From Patterns of Movement to Subjectivity of Understanding
 
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from TextCooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
 
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from TextCooperating Techniques for Extracting Conceptual Taxonomies from Text
Cooperating Techniques for Extracting Conceptual Taxonomies from Text
 
Timo Honkela: Multimodally Grounded Translation by Humans and Machines
Timo Honkela: Multimodally Grounded Translation by Humans and MachinesTimo Honkela: Multimodally Grounded Translation by Humans and Machines
Timo Honkela: Multimodally Grounded Translation by Humans and Machines
 
Using the Framework of Networks to Enhance Learning and Social Interactions
Using the Framework of Networks to Enhance Learning and Social InteractionsUsing the Framework of Networks to Enhance Learning and Social Interactions
Using the Framework of Networks to Enhance Learning and Social Interactions
 
Semiotics and conceptual modeling gv 2015
Semiotics and conceptual modeling   gv 2015Semiotics and conceptual modeling   gv 2015
Semiotics and conceptual modeling gv 2015
 
Timo Honkela: Subjects on objects in contexts: Using GICA method to quantify ...
Timo Honkela: Subjects on objects in contexts: Using GICA method to quantify ...Timo Honkela: Subjects on objects in contexts: Using GICA method to quantify ...
Timo Honkela: Subjects on objects in contexts: Using GICA method to quantify ...
 
ONTOLOGICAL MODEL FOR CHARACTER RECOGNITION BASED ON SPATIAL RELATIONS
ONTOLOGICAL MODEL FOR CHARACTER RECOGNITION BASED ON SPATIAL RELATIONSONTOLOGICAL MODEL FOR CHARACTER RECOGNITION BASED ON SPATIAL RELATIONS
ONTOLOGICAL MODEL FOR CHARACTER RECOGNITION BASED ON SPATIAL RELATIONS
 
CAL2011 Discourse-Centric Learning Analytics Briefing
CAL2011 Discourse-Centric Learning Analytics BriefingCAL2011 Discourse-Centric Learning Analytics Briefing
CAL2011 Discourse-Centric Learning Analytics Briefing
 
Ai notes
Ai notesAi notes
Ai notes
 
Natural lanaguage processing
Natural lanaguage processingNatural lanaguage processing
Natural lanaguage processing
 
XAI LANGUAGE TUTOR - A XAI-BASED LANGUAGE LEARNING CHATBOT USING ONTOLOGY AND...
XAI LANGUAGE TUTOR - A XAI-BASED LANGUAGE LEARNING CHATBOT USING ONTOLOGY AND...XAI LANGUAGE TUTOR - A XAI-BASED LANGUAGE LEARNING CHATBOT USING ONTOLOGY AND...
XAI LANGUAGE TUTOR - A XAI-BASED LANGUAGE LEARNING CHATBOT USING ONTOLOGY AND...
 
Educational Futures Evidence Hub
Educational Futures Evidence HubEducational Futures Evidence Hub
Educational Futures Evidence Hub
 
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUECOMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
 
Dimensions of Media Object Comprehensibility
Dimensions of Media Object ComprehensibilityDimensions of Media Object Comprehensibility
Dimensions of Media Object Comprehensibility
 
Listening comprehension in efl teaching
Listening comprehension in efl teachingListening comprehension in efl teaching
Listening comprehension in efl teaching
 
Listening Comprehension in EFL Teaching
Listening Comprehension in EFL TeachingListening Comprehension in EFL Teaching
Listening Comprehension in EFL Teaching
 
Machine learning-and-data-mining-19-mining-text-and-web-data
Machine learning-and-data-mining-19-mining-text-and-web-dataMachine learning-and-data-mining-19-mining-text-and-web-data
Machine learning-and-data-mining-19-mining-text-and-web-data
 
XIP Dashboard: Visual Analytics from Automated Rhetorical Parsing of Scient...
XIP Dashboard: Visual Analytics from Automated Rhetorical Parsing of Scient...XIP Dashboard: Visual Analytics from Automated Rhetorical Parsing of Scient...
XIP Dashboard: Visual Analytics from Automated Rhetorical Parsing of Scient...
 
Developing a meta language in multidisciplinary research projects-the case st...
Developing a meta language in multidisciplinary research projects-the case st...Developing a meta language in multidisciplinary research projects-the case st...
Developing a meta language in multidisciplinary research projects-the case st...
 

ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text

  • 1. Università degli studi di Bari “Aldo Moro” Dipartimento di Informatica ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella L.A.C.A.M. {fabio.leuzzi, stefano.ferilli, fulvio.rotella}@uniba.it http://lacam.di.uniba.it 9th Italian Research Conference on Digital Libraries Università la Sapienza - Rome, Italy January 31 - February 1, 2013
  • 2. Overview ● Introduction & Objectives ● Tool overview ● Knowledge Representation Formalism ● Relevant concepts ● Information Retrieval ● Reasoning by Association ● Exploiting Tool ● Conclusions & Future Works ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 2
  • 3. Introduction Some repositories leave the responsibility of quality to the authors. + Anybody can produce and distribute documents. = Possible low average quality of the repository contents. The study, understanding and exploitation of the content of a digital library, with the aim to easily explore the semantic content of huge amounts of text. ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 3
  • 4. Introduction Possible solution: ● Natural Language Processing systems ● Provide the grammatical structures contained in text ● Knowledge Representation formalisms ● Semantic networks ● Graph learning techniques ● To obtain a semantic network starting from the text ● In order to satisfy the information needs, the knowledge base can be exploited: ● To make summarizations ● To reason with it ● ... ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 4
  • 5. Objectives Improving fruition of a DL ● Use of a tool providing advanced functionalities ● Mixed strategy for relevant concept recognition ● Semantic approach to information retrieval ● Automatic inference over the acquired knowledge ● Reasoning by association ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 5
  • 6. Tool overview ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 6
  • 7. Knowledge representation formalism Only subject, verb and complement have been considered. ● Subjects and complements → concepts ● Verbs → relations between them subject, subject, verb, complement complement The frequency of arcs between the concepts in positive and negative sentences has been taken into account. ● Enrich the representation formalism ● Give robustness to our solution through a statistical approach ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 7
  • 8. Relevant Concepts ● Relevant nodes are sought in the graph ● Mixed strategy ● Semantic network structure ● EM clustering provided by Weka ● Keyword Extraction ● Quantitative approach based on co-occurrences ● Qualitative approach exploiting WordNet ● Psychological approach based on principles of an effective presentation ● Components empirically weighted ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 8
  • 9. Information Retrieval ● Word Sense Disambiguation ● One Domain per Discourse assumption: many uses of a word in a coherent portion of text tend to share the same domain ● Prevalent domain individuation ● Extraction of all synsets for each term ● Extraction of all domains for each synset ● Choice of prevalent domain synset ● Pairwise Complete Link Agglomerative Clustering ● Each synset generates a singleton cluster ● For each pair of clusters ● If the complete link property holds ● Merge the involved clusters ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 9
  • 10. Information Retrieval ● Multi-strategy Similarity Measure on WordNet ● 3 components summed and normalized in ]0,1[ ● depth (ancestors) ● breadth (direct neighbors) ● breadth (inverse neighbors) ● Document Partitioning ● For each document ● Each synset votes for a cluster ● User Query Processing ● Brute force WSD to find the best synsets combination ● Best combination used to return a ranked list of clusters ● Each cluster has a list of related documents obtained by the Document Partitioning phase ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 10
  • 11. Reasoning ‘by association’ Breadth-First Search Given two nodes (concepts), a Breadth-First Search starts from both nodes, the former searches the latter's frontier and vice versa, until the two frontiers meet. Then the path is restored going backward to the roots in both directions. We also provide the number of positive/negative instances, and the corresponding ratios over the total to help understanding different gradations (permitted, prohibited, typical, rare, etc.) of actions between two objects. ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 11
  • 12. Reasoning ‘by association’ Breadth-First Search The table below shows a sample of possible outcomes. E.g., an interpretation of case 1 can be: “the young looks television that talks about (and criticizes) facebook, because it typically does not help (rather distracts) schoolwork”. ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 12
  • 13. Reasoning ‘by association’ Probabilistic approach Real world data are typically noisy and uncertain → need for strategies that soften the classical rigid logical reasoning Defined a formalism based on ProbLog language: pi :: fi ● fi : ground literal of the form link (subject, verb, complement) ● pi : ratio between the sum of all examples for which fi holds and the sum of all possible links between subject and complement ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 13
  • 14. ConNeKTion ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 14
  • 15. ConNeKTion ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 15
  • 16. ConNeKTion ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 16
  • 17. ConNeKTion ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 17
  • 18. ConNeKTion ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 18
  • 19. Conclusions ConNeKTion allows to learn conceptual graphs from plain text and to enrich them by finding concept generalizations. The resulting graph can be used for several purposes: ● finding relationships between concepts (if any) ● filtering the concepts from a particular perspective ● relevant concepts recognition and information retrieval A suitable control panel is provided for the user to comfortably carry out these activities. ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 19
  • 20. Future Works We plan to improve the natural language text pre-processing using anaphora resolution in order to replace, where possible, pronouns with the explicit concept they express. All functionalities have parameters set empirically. A criteria for automatical setting of suitable parameters is needed. The preseted functionalities are based on the exploitation of WordNet. A strategy to make the operators WordNet free can be desirable. We also wish to extend the reasoning operators by adding an argumentation operator, that could exploit probabilistic weights, intended as a rate of reliability, to provide support or attack to a given statement. ConNeKTion: A Tool for Exploiting Conceptual Graphs Automatically Learned from Text F. Leuzzi, S. Ferilli, F. Rotella 20