Presentation is loading. Please wait.

Presentation is loading. Please wait.

Smart organization of agricultural knowledge: the example of the AGROVOC Concept Server and Agropedia ISKO Italy Open conference systems, Paradigms and.

Similar presentations


Presentation on theme: "Smart organization of agricultural knowledge: the example of the AGROVOC Concept Server and Agropedia ISKO Italy Open conference systems, Paradigms and."— Presentation transcript:

1 Smart organization of agricultural knowledge: the example of the AGROVOC Concept Server and Agropedia ISKO Italy Open conference systems, Paradigms and conceptual systems in KO Roma, 24 February 2010

2 Few words about myself

3 Outline Why such projects The AGROVOC Concept Server – Benefits – Technology The Agropedia Project – Benefits – Technology Conclusion

4 Why such projects? Adding semantics to Agricultural Knowledge – Agricultural Ontology Service Scope – Better define and describe knowledge – Give meaning and structure to information – Enable reuse of domain knowledge – Avoid ambiguities – Allow better searches – Provide smart services – …

5 The starting idea… Semantic technologies were evolving – Ontologies – Concepts – URIs – Machine readable formats Everything started from AGROVOC… – Multi-lingual – Multi-domains – Re-engineering

6 Foundational Layer Application Specific Layer Domain Specific Layer Lexicalizations Foundational Agricultural Ontology Rice Ontology Pest Ontology Plant Ontology Agricultural Domain Specific Ontology imports Indian Rice Ontology Rice Cultivation Ontology imports Application Specific Ontology imports Architecture of AOS ontologies

7 AGROVOC Concept Server

8 A knowledge base of Agricultural related concepts organized in ontological relationships (hierarchical, associative, equivalence) Will contain 600.000 terms in around 20 languages Concepts can be organized in multiple categories

9 Terminology Workbench AGROVOC OWL AOS Core: the Concept Server Export AGROVOC RDFS formats (e.g. SKOS) and TagText ISO2709 Other thesauri and terminologies integration ABACA NT1 Food NT2 Apple ANIMAL BT Organ NT.... mapping Other thesauri & terminologies ABACA NT1 Food NT2 Apple ANIMAL BT Organ NT....

10 Three levels of representation Concepts (the abstract meaning) – Ex: ‘rice’ in the sense of a plant, Terms (language-specific lexical forms) – Ex: ‘Rice’, ‘Riz’, ‘Arroz’, ‘ 稻米 ’, or ‘Paddy’ Term variants (the range of forms that can occur for each term) – Ex: ‘O. sativa’ or ‘Oryza Sativa’, ‘Organization’ or Organisation’

11 Concept example Organization – hasLexicalization Organizações ( pt ) Organization (en) [P. T] hasSpellingVariant » Organisation (uk-en) – hasSubClass department (en) – hasStatus Published – hasDateCreated 12/12/2006 – hasDateUpdated 01/10/2009

12 Semantic Relationships Concept to Concept isA (hierarchy), isPestOf, hasPest Concept to Term hasLexicalization (links concepts to their lexical realizations) Term to Term isSynonymOf, isTranslationOf, hasAcronym, hasAbbreviation Term to String hasSpellingVariant, hasSingular

13 Towards the Concept Server AGROVOC cleaning and refinement Current AGROVOC MySQL Improved AGROVOC MySQL AGROVOC OWL Revision and Refinement

14 Ontology models (AGROVOC Concept Server, LIR,...) Concept Relationships between concepts Lexicalization/ Term String Relationships between strings Relationships between terms designated by manifested as Other information: language/culture subvocabulary/scope audience type, etc. Note annotation relationship Relationship Relationships between Relationships All terms are created as instances of the class o_terms. All at the same level. Only one language per term. term level string level concept level

15 The Workbench A web-based working environment for managing the AGROVOC Concept Server Facilitate the collaborative editing of multilingual terminology and semantic concept information It includes administration and group management features It includes workflows for maintenance, validation and quality assurance of the data pool

16 Users/Roles/Groups Non registered users Term editors Ontology editors Validators Publishers Administrators

17 Modules Home Search Concept/Term Management Relationship Management Classification Scheme Management Validation Consistency Check Import/Export User/Group Management Statistics/Preferences

18 Concept/Term Management

19 Concept Relationship Can create the concept-concept relationship Inverse relationship is also created automatically Ex: If we create A affects B, then B isAffectedBy A relationship is also created

20 Graphical Visualization

21 Term Relationship Add/edit/delete term- term relationship Relationships can be – is scientific name of – has scientific name – has synonym – has translation – is acronym of – has acronym – has abbreviation

22 Term Spelling Variant Can assign the different spelling variant for the terms in different languages Ex: – color (us-en) – colour (uk-en)

23 Classification Schemes

24 RSS

25 Web services

26 System Architecture (1/2) Triple store database (MySQL and sesame) System database (MySQL) AJAX technology (Google Web Toolkit) Java Queries to the triple store using SEMRQL Organized in modules

27 System Architecture (2/2) Ontology repository (OWL) System Data Repository Protégé OWL API JDBC (MYSQL) Validation StatisticsUser Management Group Management System Preference GWT Concept Management Relationship Management SearchScheme Management ImportExportConsistency Check AGROVOC WORKBENCH CONCEPT SERVER INTERFACE

28 Benefits Agricultural related concepts will be uniquely identified – URI-based indexing and search systems Multiple terms in many languages (include spelling variants, acronyms, dialectal forms or local terms used in specific geographical area) – freedom to use any language Ability of creating catalogues more machine- interpretable; More interoperability with other systems using ontologies – mapping and linking to other URI

29 Agropedia

30 What is Agropedia Indica Knowledge Repository on Agriculture Of universal knowledge models And localized content For a variety of users With appropriate interfaces Built in collaborative mode In multiple languages RiceSaket-4 English Hindi Telegu Spanish this is a document about rice and its pests..... Once the rice ap- pear in the world..... Mad Cow Disea- se is the commonly used name for Bovine Spongiform Encephalopathy (BSE)....

31 Scope Build an infrastructure of agricultural knowledge – Multilingual and localized information – Knowledge Models (KMs) as conceptual reference – Different crops (Chickpea, Groundnut, Litchi, Pigeon pea, Rice, Sorghum, Sugarcane, Vegetable pea, and Wheat) – Domain specific information (local fertilizers, soil, cropping techniques and methods, …) Present it in various ways Different stakeholders: scientists, students, extension workers, farmers, policy makers, agronomists, soil scientists, plant breeders or geneticists, farm managers, and other experts Specific guidelines Registry of relationships (object properties and data type properties)

32 Knowledge Objects this is a document about rice and its pests..... Once the rice ap- pear in the world..... Mad Cow Disea- se is the commonly used name for Bovine Spongiform Encephalopathy (BSE).... docs, pdf, txt,... jpg, gif, bmp,... wav, audio,... htm, html, asp, php,... author:... subject:.... identifier:.... author:... subject:.... identifier:.... author:... subject:.... identifier:.... author:... subject:.... identifier:.... METADATA URI

33 Retrieval this is a document about rice and its pests..... Once the rice ap- pear in the world..... Mad Cow Disea- se is the commonly used name for Bovine Spongiform Encephalopathy (BSE).... results.....

34 Navigate knowledge maps Concept indexing Blogs (experts can create blog on specifics topics and farmers can post questions and comments) Q/A forum FAQ Agrowiki (a common platform where everyone can share experiences) Multilingual services Services

35 Knowledge base structure Agricultural Experts can upload content as: – crops calendar – publications (journals, articles, magazines, thesis, books) – do’s and don'ts (extension knowledge) – sponsor content –... Content (except agrowiki) will be verified by experts Agricultural related issues in Agrowiki

36 Conceptual Architecture Digital Objects Resource Layer Semantic Layer Interface Layer User requests Knowledge model server Upload view Content

37 Technology First release implemented using Alfresco Subsequently, because of the need of incorporating other functionalities, Drupal – blogs, chats, forums, Q/A, user management, etc. Cmap for the KMs, and exported in SVG format Other formats (pdf, jpg) for visualization only Java to customize the OWL version of the Kms Taxonomy module for tagging and searching the content A Java module for automatic tagging using an the KMs is in process of implementation.

38 Agropedia Indica Application UI Technical infrastructure Msql User management User requests Upload view Content

39 Knowledge Models A knowledge model is a function of its use For the same domain one needs multiple models depending on the use/user Researchers needed to identify these different models and build them Consistent and coherent

40 KM in Agropedia and AOS Agropedia KMs AGROVOC 30% 70% 16% of all concepts in Agropedia KM are scientific names or common names 16%

41 ENGLISH Multilinguality Generic model Specific models HINDI TELUGU.... Generic model (Specific models from IITK) translate AGROVOC Concept Server (via WS)

42 Innovative aspects Agropedia presents to users different semantically oriented tools: textual and audio blogs, wikis, forums, and the KMs presented in different formats (pdf, static or context-sensitive images) Users have the possibility to choose a preferred way of navigating the KMs Resources from the library catalogue are tagged with concepts from the KM No matter what languages the maps are displayed, the results will be always the same (currently, KMs exists in English and Hindi)

43 Agropedia What are you interested today? - FAQ - Pesticides - Rice - Seasonal info - Agroclimatic zones -.... Who are you? Agro-scientist Extension worker Call Center Operator News has Users Sponsors NAIP ICAR Partners IITK IITB ICRISAT FAO GB PANT..... has Sponsors has Partners FAQKisan Blog has Services has content..... Home About

44 Knowledge Models in Agropedia Crop Pesticides Rice – Rice pests – Rice diseases... many others

45 Crop

46 Rice cropping system

47 Rice pests

48 Rice diseases (detail)

49 Insecticides (detail)

50 Relationships concept-to-concept and instance-to-instance

51 Benefits Agropedia attempt to inject social networking and semantic technologies into Indian agriculture The Library section of the Agropedia is the expert certified knowledge Wiki, blogs, Forum provide the platform for un- regulated people-created content/knowledge Agropedia permits users to comment upon certified knowledge

52 To conclude…

53 Conclusion and Future Works FAO and AOS partners invest in processable information Agropedia opens the road to concept based maps in India A lot still remains to do, in AGROVOC CS OWL2 + knowledge extraction In Agropedia more KM + OWL for better services, e.g. Problem - solving – what should I do if my rice is infested by gundhi bug? – where I can find seeds of good quality? – what should I do if rice new leaves start yellowing? Mutual integration to investigate (same users, …) Linked Data (linkeddata.org) exposure

54 Application Specific Layer Domain Specific Layer Agropedia Indica Workbench AGROVOC CS Workbench Future AOS Ontologies Interactions IITK Modules.... rice mango sorghum.... organisms substances AGROVOC CS Modules Indian Rice Ontology@IITK Rice Ontology@IITK same URI Ecosystems Ontology@FAO May translate upper level models Agricultural Domain Specific Ontologies Other Specific Ontologies internet

55 Take home messages Semantic technologies can play a role for the agricultural domain Many stakeholders are involved (users / providers / developers) Agrovoc Concept Server and Agropedia are two project in this line

56 References http://aims.fao.org/ http://code.google.com/p/agrovoc-cs- workbench/ http://agropedia.iitk.ac.in/

57 Thanks Margherita Sini, Sachit Rajbhandari, Johannsen Gudrun, Jeetendra Singh, Johannes Keizer, Dagobert Soergel, T.V. Prabhakar, Asanee Kawtrakul


Download ppt "Smart organization of agricultural knowledge: the example of the AGROVOC Concept Server and Agropedia ISKO Italy Open conference systems, Paradigms and."

Similar presentations


Ads by Google