Keynote 1 - December 3rd, 2020 (10:00 - 11:00)
Wanida Kanarkard (Khon Kaen University)
Innovation Inspirations in Isan
The Isan region is the poorest area in the country, and life there is tough. However, challenges are what make life interesting and overcoming them is what makes life meaningful. This talk will share the experiences how to embrace the opportunities offered by the A.I, IoT, Big Data Technology to overcome these challenges and bring the better life in Isan and how they can inspire and identify new opportunities.
Keynote 2 - December 3rd, 2020 (11:00 - 12:00)
Anan Phonphoem (Kasetsart University)
GPS Tracking with LoRa: Deployment Experiences
Tracking livings is currently becoming of interest. Their positions and the corresponding timings are regularly logged within the tracking devices attached to the objects. This information can be interpreted and become meaningful to various research fields such as livestock management or health-related disease control. Moreover, with the LoRa technology, the information becomes available in near real-time for better understanding and monitoring. In this talk, the GPS tracking system with LoRa technology, developed in the IWING lab, will be presented along with the deployment experiences and lesson learns.
Keynote 3 - December 4th, 2020 (10:00 - 11:00)
Thara Angskun (Suranaree University of Technology)
Detecting Depression on Social Media using Machine Learning Techniques
Currently, depression is a problem that effects on the world population. Most of depressed people who are not be treated and represented behavior via social network posts. This research aims to develop a model of depression risk analysis from social network data. This research work collects data from a depression questionnaire with 9 questions and 499 posted opinions on social media during a two-month period after answering the depression questionnaire. These opinions reflect users’ moods, feelings, and sentiments. The proposed model applies the opinions with machine learning (ML) techniques to analyze the depression risk. There are five ML techniques explored in this research which are Support Vector Machine, Naïve Bayes, Decision Tree, Deep Learning and Random Forest. The experimental results reveal that the Random Forest technique provides higher accuracy than other ML techniques to find the depression. Moreover, the results indicate that the opinions posted on social network data such as text, emoticons and pictures can capture depressive moods of depressed users.