Activity Details

Big Data and AI in Agriculture


ObjectivesThe utilization of Big Data in agriculture is being studied globally. APAN Agriculture Working Group has conducted sessions with this title in recent years, promoting information exchange and collaboration among researchers in this field. APAN 47 will also continue this activity. Presentation such as how to make Big Data in agriculture, applications of artificial intelligence technologies to agricultural data, data management, requests from data analyst are expected.
Target AudienceMembers of Agriculture Working Group, and everyone who are interested in this topic.
Activity Co-ordinator(s)Takuji Kiura, National Agriculture and Food Research Organization (NARO), Japan
Expected No. of Participants:20
Seating ArrangementClassroom
Date:Tuesday 2019-02-19
Time:09:00 - 12:30
Location:Room 102
Session Chair(s):Veerachai Tanpipat, Hydro and Agro Informatics Institute, Thailand
Takuji Kiura, National Agriculture and Food Research Organization (NARO), Japan
No. of Participants:40, out of which 4 have provided feeedback
Agenda
1.  Introducing AgGateway Asia to promote interoperability and sustainability digital agriculture   Slides (PDF)
Rassarin Chinnachodteeraun, AgGateway Asia, Thailand
2.  Announcement of New Center and a tiny proposal   Slides (PDF)
Takuji Kiura, National Agriculture and Food Research Organization (NARO), Japan
2-1. New Center in NARO Last October, National Agriculture and Food Research Organization (NARO) established the Research Center for Agricultural Information Technology, NARO to accelerate big data and AI applications in Agriculture. . To encourage international collaboration, short introduction of new center would be reported.
2-2. Replace plan of Web APIs provided by NIAES Institute for Agro-Environmental Sciences, NARO (NIAES) provides several Web APIs to make easy retrieving of environmental open data. Unfortunately, providing Web applications could not be continued. Therefore, we are planning to replace some Web APIs with static web pages to keep data opened.
3.  A study on High Scalable Blockchain
Yuefei Gao, Department of Intelligent Interaction Technologies, China
Shin Kawai
Hajime Nobuhara
The blockchain technology, a distributed and public database of transactions, has become a platform for decentralized applications. One of these applications in agriculture is for traceability improvement. The blockchain makes it possible to track the sources of products and this helps with illness prevention and provide better lives. Despite its increasing popularity, blockchain technology still faces scalability problem. The throughput does not scale with increasing network size. When applying the blockchain technology in worldwide projects, the scalability problem may limit the performance of the blockchain. This means that blockchain may have difficulties in dealing with increasing traceability requests. Thus, in this research, we propose a scalable blockchain protocol to solve the scalability problem. The proposed protocol was designed based on proof of stake (PoS) consensus protocol and sharding protocol. The results of the experiments show that the throughput of the proposed protocol increases as the network size increase. This confirms the scalability of the proposed protocol. With the proposed blockchain, agricultural applications could be able to deal with increasing requests.
4.  Data and Workflow Management for Agriculture Map System (Agri-Map)   Slides (PDF)
Noppadon Khiripet, National Electronics and Computer Technology Center (NECTEC), Thailand
We have Thailand’s Agricultural Map for Adaptive Management (Agri-Map) operating and serving nationwide since 2016. It is an ongoing collaborative project between the Ministry of Agriculture and Cooperatives and Ministry of Science. It is a cloud-based system that allows national, regional, local policy makers and Thai farmers to access to the comprehensive, integrated and updated agricultural information through publicly available GIS applications. However, with more than 200 agricultural data layers that keep updated frequently, the workload could be burdensome for our staffs who originally manually cleansing, preprocessing and curating each data layers to be ready for the Agri-Map. Here we propose an automatic data and workflow management solution that will greatly simplify and help the updating process. This reduction in personnel working and training time will also help support the repeatability and sustainability of Agri-Map in the future.
5.  Studies and Activities for Bigdata-based Smart Farming
Masayuki Hirafuji
6.  Getting Big-Information from Big-Agricultural data: AI Methods and Potential Applications
Soumyashree Kar, Centre of Studies in Resources Engineering, Indian Institute of , India
Agricultural research domain is now more ‘deep’ than wide. While the recent developments emphasize the use of technologies and sensor-based platforms, the amount of data obtained is unarguably huge. Hence, the challenge stems not only from handling data in enormous quantities, but also in identifying the ‘useful’ portions from those datasets. Precisely, ‘how much data is informative, and what is redundant’. Hence, it’s unequivocally important to understand the statistical significance (and physical relevance) of the results obtained from transformations (as part of AI/Machine-Learning Models) performed on such huge datasets, at every step. The presentation is thus, intended to briefly focus on methods to improve precision and information from large field-phenotyping experimental designs yielding big-agricultural data, for identifying the desired crop characteristics.
7.  Data Science-based Farming Support System for Sustainable Crop Production under Climatic Change   Slides (PDF)
Seishi Ninomiya, The University of Tokyo, Japan
8.  GDP >From Agriculture   Slides (PDF)
Shah Rahman, Augere Wirless Broadband Bangladesh Ltd, Bangladesh
Bangladesh's economy grew 7.86 percent last fiscal year (2017-18) riding on the agriculture sector, especially an increase in rice production.
Summary: GDP From Agriculture in Bangladesh increased to 10468.80 BDT Million in 2018 from 10117.30 BDT Million in 2017. GDP From Agriculture in Bangladesh averaged 8879.79 BDT Million from 2006 until 2018, reaching an all time high of 10468.80 BDT Million in 2018 and a record low of 7017.10 BDT Million in 2006.
Forecast: GDP From Agriculture in Bangladesh is expected to be 10846.00 BDT Million by the end of this quarter, according to Trading Economics global macro models and analysts expectations. In the long-term, the Bangladesh GDP From Agriculture is projected to trend around 12429.00 BDT Million in 2020, according to our econometric models.
TOP