ABSTRACT:
The ontology-based data organization is one of the most popular knowledge base models thanks to its benefits of information organization and retrieval. This article is to present the agricultural data system of Tay Ninh Province which is built by using the ontology-based data organization. An ontology is a formal description of knowledge as a set of concepts within a domain and the relationships that hold between them. The agricultural data system of Tay Ninh Province has 1,269 classes, 22,612 instances, and 151 relationships among agricultural concepts and terms. This system is expected to help local farmers get agricultural information quickly and accurately.
Keywords: Ontology, agriculture data, Tay Ninh Province, technology.
1. Introduction
John Naisbitt, who warned about the booming of information technology industry, made a renowned quote “We are drowning in information but starved for knowledge”. Why? Since the introduction of World Wide Web, human knowledge has changed dramatically. As long as a smartphone, a laptop, or a desktop computer is connected to the Internet, it is very easy to find human knowledge. Maybe within a few years, human knowledge may be doubled.
To overcome this situation, Tim Berners-Lee introduced semantic web technology. "Semantic Web is the extension of current Web in which information is clearly defined so that people and computers can work together more effectively" (Tim Berners-Lee)[1]. Semantic Web is merely the extension of current Web, but not a breakthrough in replacing the old Web technology. In contrast, Semantic Web inherits from current Web and allows exploiting current Web in a new path, in which machines and people can work collaboratively in exploiting Web resources. (Christian Bizer & Tom Heath and Berners-Lee, 2007) [2]
2. What is ontology?
Ontology is a term borrowed from philosophy which is derived from Greek words "onto" and "logia". In computer science, Ontology refers to a description of concepts and the relationships of those concepts in order to represent a view of a sector or an application domain. However, a definition of ontology mentioned by Tom Gruber in the article “Toward Principles for the Design of Ontologies Used for Knowledge Sharing” is commonly used, “ontology is a description (like a formal specification of a program) of the concepts and relationships that can exist for an agent or a community of agent”. (Thomas R. Gruber) [8].
In summary, Ontology is the expression of a set of concepts (objects), in a specific domain, and the relationships between these concepts. Ontology for a sector will clearly describe the entities that help people and machines understand and deduce the meaning within that sector. Main components of Ontology are: Class, Property, and Individual. Class is a set of entities, which describes the logic to define objects of the class. Class is built according to the parent hierarchy as a classification of objects. Property represents the binary relationship of individuals (relations between two individuals) such as linking two entities together. Individual component may also be referred as “expression.” (B. Bhat & P. Bhat 2012) [1]
Figure 1. Some classes, instances, and relations among them in the wine domain (Natalya F. Noy & Deborah L. McGuinness) [5]
3. The role of Ontology in agricultural information management system of a number of non-governmental groups and organizations
Knowledge is the collection of information that is explicitly stated. Knowledge representation is a method of coding knowledge so that it can be processed by computers. Therefore, knowledge representation plays a very important role in affirming the ability to solve problems; Ontology plays a very important role in the process of creating and managing knowledge. In recent years, Ontology has become one of top topics of interest to scientific research community. Many ontologies are about a certain area of particular interest and are used as a search engine on web with the goal of making web-based search become significantly effective, especially to result in more accurate search results for domains than searching with regular keywords. Therefore, Ontology is widely used in the construction of agricultural information and knowledge databases, classification of agricultural information, and especially research and development of smart search engines as well as the implementation of cooperative information services… in some non-governmental organizations and groups specializing in agricultural information management in Vietnam and around the world (Siquan Hu & Haiou Wang & Chundong She & Junfeng Wang, 2010) [7].
a. FAO and AIMS
FAO[2] is the first professional organization of the United Nations, which was founded on October 16, 1945 at a conference in Quebec City (Canada). Since 1981, this date has become World Food Day. FAO has a set of multilingual dictionary (23 languages) named Agrovoc, which covers the following areas: food, nutrition, agriculture, forestry, and fishery… Agrovoc is maintained and developed by the editorial community with Vocbench tool, an ontology builder based on open-source web platform and strongly developed thanks to regulatory standards such as KOS, SKOS, and so on. Especially FAO has an agricultural portal developed by AIMS (Agricultural Information Management Standards) as the means to access and discuss the standards, tools, and methods of agricultural information management that connect information workers around the world to build a global practice community for agricultural information.
b. Agricultural Ontology Service (AOS)
Agricultural Ontology Service (AOS)[3] is the solution to create a knowledge organization framework in the field of food and agriculture. The purpose of AOS is to achieve the harmony between agricultural systems. Currently, AOS has developed to become a global agricultural information project that has been developed very strongly by AOS community with 13 seminars on food and agriculture.
c. Crop Ontology (CO)
Crop ontology (CO)[4] is the project established by Generation Challenge Program (GCP). This project is intended to build Crop ontology tool to support users in creating and accessing agricultural information about plants in a quick and accurate way. Current goal of Crop Ontology (CO) is to develop proven concepts along with the anatomical, structural and phenotypic relationships associated with plants.
Figure 2. Crop attribute classes (Gelian Song & Maohua Wang & Xiao Ying & Rui Yang & Binyun Zhang, 2012) [4]
4. The Ontology-based agricultural information management system of Tay Ninh Province
a. Process of building the agricultural data ontology in Tay Ninh Province
- Step 1. Determine the field and scope of the ontology: The design of plant seed ontology for Tay Ninh Province aims to help farmers, scientists and managers better understand the plant varieties in the province through the management, editing, search system, etc. The area to be built is the cultivar, including: industrial crops, food crops, crops, etc.;
- Step 2. Consider reusing existing ontologies: Currently, there are many ontologies developed by research groups, organizations, scientists… Here, we use the agrovoc ontology of the Food and Agriculture Organization of the United Nations (FAO). Ontology agrovoc is a very popular ontology for many ontology development organizations and communities for many fields and especially in the agricultural sector. Ontology agrovoc has 890 concepts (concepts), 21571 terms (Term) and 101 relationships (relationship) available;
- Step 3. List important terms: The abundance of cultivated land leads to a variety of plant varieties. We rely on the agricultural materials of farmers, plant breeding centers, plant protection centers, agricultural extension centers of the province and the Department of Agriculture and Rural Development in Tay Ninh Province has recorded 1011 agricultural terminology;
- Step 4. Identify classes and class hierarchies: From the terms listed in step 3, we have implemented class hierarchies for terms;
- Step 5. Identify attributes and relationships: After completing the class hierarchy we proceeded to build relationships and create properties for the hierarchy classes. Ontology agrovoc has 101 relationships (relationship) available. From that foundation, we built 40 more relationships between classes;
- Step 6. Determine the constraints of attributes: We conducted a check on the bond between classes. Specific expressions are displayed during the process of building relationships between classes;
- Step 7. Create instances (or entities): We performed the transfer of the remaining terms after completing step 3 into the built classes.
b. Result
Ontology-based agricultural information management system of Tay Ninh Province (hereinafter referred to as the system) is a system built in the end of 2013 based on Agrovoc dictionary of FAO by using Vobench[5] 1.3.1, an ontology development support tool, combined with the development of SKOS standards[6]. In the process of building additional concepts, terms and relationships between agricultural concepts and terms, in the end of 2015, the system had 899 concepts, 21571 terms, and 101 relationships between concepts and agricultural terms. According to the statistics of the system as of August 2017, the system has 1269 concepts, 22612 terms, and 151 relationships between agricultural concepts and terms, with an addition of 370 concepts, 1011 terms, and 40 relationships between agricultural concepts and terms.
Figure 3. The hierarchy of relationships
In the process of building additional concepts, terms and relationships between agricultural concepts and terms. By the end of 2015, the system had 899 concepts, 21571 terms, and 101 relationships between concepts and agricultural terms.
Figure 4. The decentralization of short-term industrial tree concepts
and short-term rice varieties
According to the statistics of the system as of August 2017, the system has 1,269 concepts, 22,612 terms, and 151 relationships among agricultural concepts and terms, with an addition of 370 concepts, 1011 terms, and 40 relationships among agricultural concepts and terms.
Figure 5. The hierarchy of relationships
5. Conclusions
Tay Ninh Province has abundant land potential, in which over 96% of land resource is favorable for developing all kinds of crops. Land resources in Tay Ninh Province can be divided into five main soil groups with 15 different soil types. The abundance of cultivation land resources leads to a variety of plants and crops… Therefore, the issue of building and implementing an ontology-based agricultural information management system is an important and urgent issue to serve the managers and farmers in the province to contribute to the construction, development, and integration of the province toward a smart city in the future.
In order to develop public services to assist farmers, managers and scientists in managing and finding agricultural information, the agricultural sector ontology in Tay Ninh Province will be the knowledge base for the management system. Knowledge-based agricultural information is located on the website of the Department of Agriculture of Tay Ninh Province.
FOOTNOTES:
[1] Kim Kha (2009). Sự ra đời của web ngữ nghĩa. <http://www.kimkha.com/2009/11/su-ra-oi-cua-web-ngu-nghia.html>
[2] AGROVOC tesauro multilingüe de agricultura <http://aims.fao.org/es/agrovoc>
[3] Agricultural Ontology Services (AOS). <http://aims.fao.org/type-content/agricultural-ontology-services-aos>
[4] Crop Ontology. <https://www.cropontology.org/>
[5] VOCBENCH Setup. <https://aims-fao.atlassian.net/wiki/spaces/VB/pages/1507390/Installation>
[6] SKOS home. <https://www.w3.org/2004/02/skos/
REFERENCES:
- Bhat and P. Bhat.(2012). Domain specific ontology, extractor for Indian languages. Proceedings of the 10th Workshop on Asian Language Resources, pages 75-84, COLING 2012, Mumbai.
- Christian Bizer, Tom Heath and Berners-Lee. (2009). Linked Data - The Story So Far. International Journal on Semantic Web and Information Systems (IJSWIS).
- Emmanuel Ukpe. (2013) .Agriculture Ontology for Sustainable Development in Nigeria. IOSR Journal of Computer Engineering (IOSR-JCE), e-ISSN: 2278-0661, p- ISSN: 2278-8727, Volume 14, Issue 5, pages 57-59.
- Gelian Song and Maohua Wang and Xiao Ying and Rui Yang and Binyun Zhan. (2012). Study on precision agriculture knowledge presentation with ontology. AASRI Conference on Modelling, Identification and Control, AASRI Procedia, Volume 3, Pages 732-738,
- Natalya F. Noy and Deborah L. McGuinness. (2000). Ontology Development 101: A Guide to Creating Your First Ontology. Stanford, CA: Stanford University.
- Papajorgji and P. Pardalos. (2008). Advances in modeling agricultural systems. Springer US.
- Siquan Hu and Haiou Wang and Chundong She and Junfeng Wang. (2010). Ontology for Agriculture Internet of Things. 4th Conference on Computer and Computing Technologies in Agriculture (CCTA), Nanchang (China), Oct 2010 , pages 131-137.
- Thomas R. Gruber. (1995). Toward Principles for the Design of Ontologies Used for Knowledge Sharing. International Journal of Human-Computer Studies, Volume 43, Issues 5–6, Pages 907-928.
QUẢN LÝ HỆ THỐNG DỮ LIỆU NÔNG NGHIỆP DỰA TRÊN ONTOLOGY
Ở TỈNH TÂY NINH
ThS. PHẠM NGUYỄN HUY PHƯƠNG
Khoa Công nghệ thông tin, Trường Đại học Công nghiệp Thực phẩm TPHCM
ThS. NGUYỄN THỊ BÍCH NGÂN
Khoa Công nghệ thông tin, Trường Đại học Công nghiệp Thực phẩm TPHCM
ThS. TRẦN PHONG VŨ
Khoa Công nghệ, Trường Cao đẳng Sư phạm Tây Ninh
TÓM TẮT:
Ontology là một trong số các mô hình cơ sở tri thức phổ biến nhất vì những lợi ích trong việc tổ chức và truy xuất thông tin. Vì những hiệu quả của việc tổ chức dữ liệu dựa trên ontology, bài viết xây dựng hệ thống dữ liệu nông nghiệp của tỉnh Tây Ninh theo phương thức này. Ontology là biểu hiện một tập các khái niệm (đối tượng) được gọi là các lớp hoặc các đối tượng cụ thể, thuộc một lĩnh vực cụ thể và chứa những mối quan hệ giữa các loại khái niệm này. Hệ thống cơ sở khoa học dữ liệu nông nghiệp này hiện có 1269 lớp, 22612 đối tượng cụ thể (thực thể) và 151 mối quan hệ giữa các khái niệm và thực thể nông nghiệp, nhằm phục vụ người nông dân tỉnh nhà cập nhật và nắm bắt thông tin nông nghiệp một cách nhanh chóng và chính xác dựa trên công nghệ ontology.
Từ khóa: Ontology, nông nghiệp dữ liệu, Tây Ninh, công nghệ.
[Tạp chí Công Thương - Các kết quả nghiên cứu khoa học và ứng dụng công nghệ,
Số 18, tháng 7 năm 2020]
