Ontology-driven crop base knowledge system

CFFRCPLUS CropB1-001

Level: PhD
Value: MYR 226,535
Principal investigator: Abdur Rakib, School of Computer Science, UNMC (Abdur.Rakib@nottingham.edu.my)
Co-investigator: Natasha Alechina, School of Computer Science, UoN (Natasha.Alechina@nottingham.ac.uk)

Summary

Diversifying crops has been recognised as an important factor in reducing starvation and malnutrition. Furthermore, it plays a crucial role in increasing output of agricultural economy. The potential of underutilised crops, such as, for example, Bambara groundnut has been realised to some degree [1]. In most of the African countries, it is considered to be the third most important legume after groundnut and cowpea [2]. Even though there has been a considerable amount of research in increasing the productivity of Bambara groundnuts and related topic [3] [4], however, very little attention has been directed towards developing formal knowledge bases that facilitate sharing data and knowledge among disparate researchers. This project will develop an ontology-driven crop base knowledge system tool to address this gap. In particular this research will contribute to Knowledgebase Population and Continued Platform Development in the CropBase. We use ontology-based technique, because it is an explicit formal specification of a conceptualization which defines certain terms of a domain and the relationships among them [5]. It provides a formal representation of knowledge as a set of concepts within a domain, and it can also be exploited to reason about the entities within that domain. Furthermore, it provides the semantics necessary for accurate and unambiguous integration of data from disparate sources, as an example, the integration of data about the crop breeder, crop user, affecting diseases etc. It is part of a multidisciplinary research project, crop related data required for this project will be obtained from other collaborative CFFRC research projects.

Background

There is a wealth of information available in different sources about crops, in particular about underutilised crops such as Bambara groundnuts, their genetics and agronomy. Some of this information is available in informal sources on the web, for example on Wikipedia [6]. However, this knowledge is not entirely authoritative, it may be incomplete and for these reasons it is not suitable as a basis of decision making by crop growers. Moreover, there is little research has been carried out for developing formal web-based knowledge system for underutilised crops that aims: (1) to eliminate inconsistencies in terminology (2) improper syntax and semantics, those are the main obstacles to sharing data and knowledge among disparate researchers. There are formal plant and crop ontologies, for example [7]. However, such ontologies tend to concentrate on main stream and not on underutilised crops. The aim of this project is to combine and expand existing crop ontologies and if needed modify them to ensure consistency, include data for underutilised crops, and provide support for decision making by crop growers in selecting suitable crops for their circumstances, including the type of land and climate, likely plant diseases, pests and other problems, and implement a web tool to provide this support with an ontology in the background.

Problem statement

The core problem which this study intends to investigate is three-fold:

  1. We need to identify knowledge gaps. What are the factors important for adopting crops? In particular, why are some crops with good economic potential underutilised? There could be many different information sources that will help us understand the problem clearly, such as, for example, knowledge about geographical area, social knowledge, and knowledge about economic impact. The required data may be obtained from other collaborative CFFRC research projects

  2. We need to populate the knowledge base combining existing information sources and if necessary formalising new knowledge.

  3. We need to develop a knowledge tool which uses the knowledge base which can answer queries by researchers and crop growers to assist in their decision making for choosing crops. Sharing and integration of knowledge among different sub-communities (CropFinder, CropMapper, CropBreeder, CropGrower, and CropUser) by establishing a common ontology for integrating their diverse data sources.

Objectives

In this research work, we would like to develop an ontology-driven CropBase knowledge system for underutilised crop research and development. It will contribute to Knowledgebase Population and Continued Platform Development. Research activities will include design and implementation of an ontology-driven integration tool that will enable access to information and assist in decision making for crop growers.

Methodology

To achieve goals of the proposed research, we apply the following methodology:

  1. Studying existing information: analysing existing standard ontologies in the domain of Crops (focusing on underutilised crop, e.g., Bambara groundnut) and identify potential problems and semantic ambiguities. Identify and collect information sources required for the development of the knowledge tool.

  2. CropBase ontology design and development: combine existing crop ontologies and expand them with new information concerning underutilised crops, crop diseases and other relevant knowledge. Express this information in Web Ontology Language (OWL) which extends RDF and which has been developed as an ontology language that defines classes and properties and their relationships. The World Wide Web Consortium (W3C) has declared two different standards for OWL, namely, OWL 1 and OWL 2. We will use OWL 2 DL, which is based on description logic (DL), a decidable fragment of first order logic that is used for efficient and tractable reasoning. We will use a state-of-the-art ontology editor and knowledge-base framework such as e.g., Protege. for authoring the ontology.

  3. CropBase tool: the tool will use the ontology as the main knowledge source for answering user queries, but will also involve an additional rule-based decision support mechanism which will ask for information about the concrete parameters relevant for the crop grower and display information about suitable crops in the form which is precise but easy to understand for non-computer scientist. In technical terms, the tool is a hybrid system which uses both ontology definitions and rules, based on the principles in for example [8].

Evaluation

We will evaluate scalability and expressiveness of our approach by using a case study, which will determine the relationship between the size of a knowledge base and time required by the tool to answer typical queries. Underutilised crop data required for this project will be obtained from other collaborative CFFRC research projects.

Ideal candidate

The successful candidate will have a strong background in computer science and must have excellent programming skills (particularly Java), a preliminary knowledge of ontology. Some knowledge of, or interest in, rule-based reasoning and/or AI techniques, would be desirable. The successful candidate will also collaborate with other CFFRC researchers.

Application

All applicants should read the entry requirements and eligibility criteria and applications must include:
(1) A complete CV
(2) One page research statement indicating what you see are interesting research issues relating to the above post and why your expertise is relevant.
(3) 2 References
(4) English language Applicants must have one of the following English qualifications:

  1. 0 level/GCSE English: grade A to C

  2. JMB Test: grades 1, 2, s

  3. Test of English as a Foreign Language (TOEFL): minimum mark of 550

  4. IELTS: minimum score of 6.0

Closing Date: 10th February 2013

References

[1] Azam-­Ali, S. N., Sesay, A., Karikari, S.K., Massawe, F. J., Aguilar­-Manjarrez, J., Bannayan, M., and Hampson, K.J., (2002) Assessing the potential of an underutilized crop-A case study using Bambara groundnut. In Experimental Agriculture, Volume 37, Issue 04, pp 433-472.

[2] Hampson, K., Sesay, A., Mukwaya, S.M., Azam-Ali, S.N. (2000) Assessing opportunities for increased utilisation of bambara groundnut in Southern Africa. Tropical Crops Research Unit, School of Biosciences, University of Nottingham. DFID CPHP Project R-7527.

[3] European Union FP-5 INCO-DC (2001). Increasing the Productivity of Bambara Groundnut (Vigna subterranea (L.) Verdc) for Suistanable Food Production in Semi-Arid Africa. Report 1, 68 pages.

[4] Azam-Ali, S.N., Aquilar-Manjarrez, P and Bannayan, M. (2001). A Global Mapping System for Bambara Groundnut, FAO Monograph. 56 pages (FAO, Rome).

[5] Gruber, T.: A translation approach to portable ontology specifications (1993). Knowledge Acquisition- Special issue: Current issues in knowledge modeling 5(2) 199-220.

[6] http://en.wikipedia.org/wiki/Bambara_groundnut

[7] http://www.cropontology.org/

[8] Motik, B., Sattler, U., and Studer, R. (2005). Query Answering for OWL-DL with rules. J. Web Sem. 3(1): 41-60 (2005).