Lương Văn Thiện is a PhD student at Queen's University Belfast, the UK . Photo courtesy of Thiện

Lương Văn Thiện, PhD, is the youngest among the top 10 scientists to receive the 2022 Golden Globe Science-Technology Award. His AI solution 'Anomaly Detection for Time-series' has been applied in telecommunications troubleshooting. Thanh Nga talks with him about his journey.

Inner Sanctum: Please tell us a little about yourself.

Hi, I'm Lương Văn Thiện, born in 1992. Now I am a lecturer at the Faculty of Computer Science, head of AIoT Lab, Phenikaa University in Hà Nội. I studied at the Faculty of Electronics and Telecommunications of the University of Science and Technology Hà Nội (USTH).

In 2019, I completed my PhD thesis at Queen's University Belfast (QUB), the UK. I was honoured to become one of 10 young scientists to receive the 2022 Golden Globe Science-Technology Award (organized by the Ministry of Science and Technology and the Central Committee of the Hồ Chí Minh Communist Youth Union).

Inner Sanctum: You received a full scholarship for doing a PhD in the UK without doing a master's? What helped you get this?

The first time I practised scientific research in my final year at USTH in 2015, it was under the guidance of Dr Ngô Vũ Đức of USTH. He gave me a very difficult maths problem, but my background knowledge of the topic was limited. I had to read many documents and books in English to better understand what I would do.

Thanks to my efforts, I solved the difficult problem and the result was that my two subjects were accepted at two prestigious international scientific conferences, including the IEEE PIMRC 2015 and IEEE ATC 2015.

As a result, my scientific research ability was proved through these two works, leading to a full PhD scholarship at QUB in 2016 without doing a master's.

Now, I believe that, if you persist in pursuing the right goal with a clear mind, sooner or later you will achieve your aim.

Lương Văn Thiện (centre) celebrates winning the 2022 Golden Globe Science-Technology Award. — Photo courtesy of Thiện

Inner Sanctum: After graduating in the UK, you had many opportunities to stay back for work but decided to return to Việt Nam. Why?

I was born and raised in a poor rural area in Ninh Bình Province. My parents were farmers. They had to work hard to earn money to feed their children.

At that time, I did not have money to take private lessons, and my family did not have an academic tradition, so I searched for advanced maths books. I brought the books with me to study, while I tended cows; so when I finished the 6th grade, I had done all the maths exercises of grade seven.

I share my story not to complain about hardship, because I'm sure many other students had similar situations like me, they all had the desire to learn and rise out of poverty when witnessing the hard work of their parents.

Therefore, when I received the opportunity to study abroad, I decided to study and return. I wanted to return to inspire and teach the students because I saw in them "myself" in the past. Through every hour of teaching, I share with students the importance and responsibility of studying to bring value to themselves and the society.

Lương Văn Thiện, PhD, (centre) hopes to develop AIoT Lab into a strong research group, together with students, creating many influential scientific works, useful technological solutions. — Photo courtesy of Thiện

Inner Sanctum: The AI solution for 'Anomaly Detection for Time-series' (ADT) is being practically applied by Mobifone Telecommunications Equipment Testing and Repair Center to detect abnormal changes in network and transmission quality. Can you tell me more about this?

This ADT solution helps to quickly detect Base Transceiver Stations, mobile cells that have abnormal changes in network quality (such as decrease in download and upload) and transmission quality (such as increase package loss, jitter and delay).

From these findings, Mobifone can provide timely warnings, combine correlation analysis with other technical data sources, assist engineers in determining the cause of abnormalities as well as provide solutions, thereby helping Mobifone quickly fix problems, ensuring customers have the best network quality.

Before this solution, it was very difficult for Mobifone engineers to monitor a considerable number of mobile stations and cells, not to mention find the causes for the problems. It took a lot of time and effort. The ADT solution is not only more effective by fully exploiting telecommunications data, but also significantly reduces pressure on engineers.

Inner Sanctum: Can you tell me a little bit about your other research projects?

I would like to mention the time and attendance system with face recognition, FaceRA. This has been deployed to enrol students at AIoT Lab, Phenikaa University with high accuracy, with anti-spoofing such as using photos or videos with faces to take attendance.

I also researched a solution about data analytics for Vietnamese immune systems. Thanks to these important indicators, doctors can quickly and accurately diagnose the patient's condition to come up with the most appropriate treatment plan, and avoid wasting time and resources by having to test too many indicators.

Inner Sanctum: What inspired you to do such practical research?

Every day, hundreds of research works in IT are published in domestic and international scientific magazines, so it will be difficult to capture all that information.

I have to read a lot of public white papers in technology forums to develop the ideas, and find the right direction to follow. Moreover, the most important basis is the need to have an accurate vision and good background knowledge to stick to the reality of the country to avoid creating something society does not need. I always think that making the right choice is more important than the actual effort, so I am careful about each choice of research ideas.

Lương Văn Thiện (left) seen at the recent scientific conference. — Photo courtesy of Thiện

Inner Sanctum: Can you tell about the application of AI in general?

AI helps create significant changes in education. Educational activities such as marking or tutoring students can be automated thanks to AI technology. Many games and educational software were born to meet the specific needs of each student, helping improve learning.

AI can also point to issues that need to be improved. For example, when many students are found to send in the wrong answers to an exercise, the system will notify the teacher and send a message to the student to correct the answer. AI technology also can monitor student progress and notify teachers.

Furthermore, students can learn anywhere in the world through the use of AI-powered software. AI technology also provides data to help students choose the best courses for them

Inner Sanctum: What are your plans for the future?

As a lecturer at the Faculty of Computer Science at Phenikaa University, I want to inspire students directly. Doing research is difficult as it needs time and effort. I think perseverance and belief will help students pursue their goals.

I hope to develop AIoT Lab into a strong research group, together with students, creating many influential scientific works, useful technological solutions, while helping to solve real-life problems with the focus on AI and IoT technologies. I want to help students become experts in their field, not only good in their profession, but also beautiful in character. VNS