Transcript Phase 1
• Context • Problem • Research Question A Framework to Experiment with Different NLP Techniques • Background • Framework Ricardo Gacitua1, Pete Sawyer1, Paul Rayson1, Scott Piao2 1 Computing 2 Department, Lancaster University, Lancaster, UK School of Computer Science, Manchester University, U • Results • Demo • Conclusions • Further Work Workshop - Issues in Ontology Development and Use Nottingham, UK. 2007 Index • Context • Problem Context • Research Question Problems • Background • Framework • Results • Demo • Conclusions • Further Work Research Question Objectives Framework Brief Demo – Ontolancs –Workbench Further Work Context Focus: • Context Most initiatives for Ontology Learning combine techniques to find concepts and relationships between them. • Problem • Research Question Extracting the relevant domain terminology and synonyms from a text • Background collection Discovering concepts which can • Framework • Results • Demo • Conclusions • Further Work Deriving a concept hierarchy be regarded as abstractions of human thought Methods for term extraction can organizing be as these concepts simple as : Unsupervised clustering techniques Extending an existing concept •counting raw frequency, hierarchy with new concepts Learning taxonomic relations known from Machine Learning. •applying information retrieval between concepts [Cimmiano et al. 2005, faure & methods such as TFIDF (Baeza-Yates & Learning non-taxonomic Nedellec 1999, Caraballo, 1999] relations between concepts Ribeiro-neto, 1999) or Populating the ontology with instances of relations and concepts •applying sophisticated methods such as the C-value / NC-value method Discovering other axiomatic [Frantzi & Ananiadou 1999] relationships or rules involving concepts and relations. Context • Context • Problem • Research Question Focus: Most initiatives for Ontology Learning combine techniques to find concepts and relationships between them. • Background • Framework • Results • Demo However, researchers have realised that the output for the ontology learning process is far from being perfect [Cimmiano, 2005] • Conclusions • Further Work Philipp Cimiano, Johanna Völker, Rudi Studer Ontologies on Demand? - A Description of the State-of-the-Art, Applications, Challenges and Trends for Ontology Learning from Text Information, Wissenschaft und Praxis 57 (6-7): 315-320. October 2006. see the special issue for more contributions related to the Semantic Web Problem • Context A key issue not addressed yet: In most cases, it is not obvious • Problem to how to use, configure and combine techniques from • Research different fields for a specific Question domain. A challenging issue is to quantitatively evaluate the usefulness, accuracy of the techniques and combinations of techniques when applied to ontology learning [1]. • Background • Framework • Results • Demo • Conclusions • Further Work Reinberg and Spyns (2005) point out the importance of the evaluation of the effectiveness of the techniques for ontology learning “To our knowledge no comparative study has been published yet on the efficiency and effectiveness of the various techniques applied to ontology learning”. (page 2) (1) Reinberger, M. L. and P. Spyns (2005). Unsupervised text Mining for the learning of DOGMA-inspired Ontologies. Ontologies Learning from Text: methods, Evaluation and Applications, Advances in Artificial Intelligence. P. Buitelaar, Cimiano P., Magnini B. (eds.). Amsterdam, IOS Press. vol. 24,: pages 305-339. Research Question • Context • Problem • Research Question • Background • Framework • Results • Demo • Conclusions • Further Work Can shallow semantic analysis of the kind enabled by semantic tagging, together with a range of other statistical NLP techniques; identify key domain concepts? Can it do it with sufficient confidence in the correctness and completeness of the result?” Background.. A number of frameworks that support ontology learning process have been reported: • Context • Problem • Research Question • Background •Background • Framework They implement several techniques from different fields such a knowledge acquisition, machine learning, information retrieval, natural language Most frameworks use a pre- processing, artificial intelligence reasoning and defined combination of database management. techniques. Thus, they do not include any mechanism for carrying out experiments with combinations or the ability to include new ones. • Results • Demo • Conclusions • Further Work ASIUM Tex2Onto Text2Onto is based on the GATE framework. GATE framework it is flexible with respect to the set of algorithms. OntoLearn OntoLT DODDLE A Flexible Framework Phase 1: Part-of-Speech (POS) and Semantic annotation of corpus: •Domain Context texts are tagged morphosyntactically and semantically. • Problem Phase 4: Domain Ontology Edition: the bootstrap ontology is turned into OWL. Then it is processed using an ontology editor (Protégé) to manage the versioning of the domain ontology and modify or improve it. • Research Question • Background A existing DAML ontology can be used as a reference and to calculate precision and recall. • Framework Phase 2: Extraction of The domain • concepts: Results terminology is extracted from the • tagged Demo domain corpus by identifying a list of domain candidate terms. The system • provides Conclusions a set of statistical and linguistic techniques which an • ontology Further Work engineer can combine •Phase 3: Domain Ontology Construction: Concepts extracted during the previous phase are then added to a concept hierarchy. Preliminary Results • Context • Problem • Research Question Some researchers use different text processing techniques such as stopword filtering, lemmatization or stemming. StopWord Filtering: [ Bloehdorn et al., 2006 ] Lemmatization: [ Buitelaar and Ramaka, 2005 ] Stemming: [ Kietz et al, 2000 ] •S. Bloehdorn and P. Cimiano and A. Hotho: Learning Ontologies to Improve Text Clustering and Classification. Proc of GFKL, 2005. •Paul Buitelaar, Srikanth Ramaka Unsupervised Ontology-based Semantic Tagging for Knowledge Markup In: Proc. of the Workshop on Learning in Web Search at the International Conference on Machine Learning, Bonn, Germany, August 2005. •J.Kietz, et al., A Method for semi-automatic ontology acquisition from a corporate intranet, in: Proc EKAW-2000 , France. 2000. • Background • Framework •• Results Results From the preliminary experiments, we can conclude that the lemmatization technique (Group 3) produces better results than the stemming technique (Group 2) for the domain concept acquisition process. • Demo • Conclusions • Further Work Our results are consistent with other studies. For instance, Alkula[3] suggests that the lemmatization may be a better approach than stemming. [3]Alkula, R. 2001. From Plain Character Strings to Meaningful Words: Producing Better Full Text Databases for Inflectional and Compounding Languages with Morphological Analysis Software. Inf. Retr. 4, 3-4 (Sep. 2001), 195-208. Brief Demo • Context • Problem • Research Question • Background • Framework • Results •• Demo Demo • Conclusions • Further Work Ontology Framework Conclusions Main challenge: • Context Our research project addresses an important challenge of ontology research, i.e. how quantitatively to evaluate the usefulness and accuracy of both techniques and combinations of techniques, when are applied to ontology learning. • Problem • Research Question • Background • Framework 1 This framework is designed as a cyclical process to experiment with different techniques. Techniques are included as a plug-in. 2 It provides support to determine what techniques or their combination provide optimal performances for ontology learning • Results • Demo • Conclusions • Further Work Our ontology learning environment in unique in not only providing a framework for integrating linguistic techniques, but also possibility an experimental platform for identifying the most effective technique or combinations. Further Work Our Project: • Context • Problem OntoLancs – A Flexible Framework For Ontology Learning • Research Question • Background • Framework Including new techniques (plugin) from different tools. Experimenting with techniques in a Supervised and Unsupervised Mode • Results • Demo • Conclusions •• Further Further Work Work Future Work A graphical workflow engine will provide support for the composition of complex ensemble techniques Integration with Protégé (Editor) The End • Context • Problem • Research Question • Background • Framework • Results OntoLancs • Demo Computing Department • Conclusions • Further Work Lancaster University 2006, UK Text2Onto vs. OntoLancs • Context • Problem • Research Question • Background Text2Onto defines the user interaction as a core aspect whereas our framework provides support to process algorithms in a unsupervised mode. Our framework provides a graphical workflow engine to provide support for the composition of complex ensemble techniques. • Framework • Results • Demo • Conclusions • Further Work Our framework uses a plug-in-based structure as Text2Onto. However, in contrast, it can include techniques from existing linguistic and ontology tools by using java API’s. Techniques included into OntoLancs • Context • Problem • Research Question • Background • Framework • Results • Demo • Conclusions • Further Work 1. 2. 3. 4. 5. 6. 7. 8. Grouping by POS Raw Frequency Filtering POS Filtering Lemmatization Stemming StopWord Filtering Frequency Profiling Syntactic Pattern Coocurrences 9. Window-based Collocations 10. Semantic Filter (soon)