Freehand China·Explanation of "Words"丨Use big data to decipher the Chinese character code to make "unpopular secrets" popular

Dahe News Everyone has heard of Oracle, but do you know Oracle? How many oracles are there? What do they all look like? As an ancient text, oracle bone inscriptions are far away from the general public. The activation and utilization of oracle bone inscriptions is a big issue that we currently need to study and solve.

In 2019, Liu Yongge, a teacher at the School of Computer and Information Engineering of Anyang Normal University, led a team to develop the world's first free Oracle big data platform "Yin Qi Wenyuan", which made a stunning debut, bringing Oracle closer to the general public.

How will a computer understand the text written 3,000 years ago? Oracle "hits" modern technology, how to keep it from being "unpopular"? How to show the charm of Oracle in a form that is popular with the public?

Oracle, which has been sleeping for 3,000 years, urgently needs to be "awakened"

"In 2000, when I returned to school, two teachers who studied Oracle strongly urged me to carry out Oracle informatization research, but my understanding of Oracle was equivalent to blank." As an undergraduate majoring in mathematics and a recent graduate with a master's degree in computer software, Liu Yongge, From that moment on, I decided to take Oracle informatization research as my lifelong career.

When talking about his research work, Liu Yongge jokingly described it as "blocked at both ends." He said: "One end is the oracle bone inscriptions of China's oldest civilization, and the other end is advanced computer technology and artificial intelligence."

Oracle bone inscriptions were discovered in the north of Xiaotun Village, Wuli, northwest of present-day Anyang City, more than 3,000 years ago. This ancient writing is also full of mystery and unknown due to its age. Decoding the oracle bone inscriptions has become a difficult problem in the world, and there is still another problem to crack. In front of us - we have no way to input Oracle into the computer.

"Because oracle bone script is different from Chinese characters. We don't recognize many of its characters, so you can't use the pinyin input method. Some characters are like a painting, so you can't use radicals or split them. Previously, the oracle bone script input method Using coded input results in greater memory burden and learning cost," Liu Yongge said.

Liu Yongge said that in the past, whether it was writing papers, giving lectures, or doing research, all situations required pasting pictures. It was crucial to develop an Oracle input method and establish a font library so that anyone can input easily.

"Using computer technology to study Oracle's encoding, fonts, input methods, and database construction, and providing some tools for Oracle experts to assist in Oracle research, are the main tasks of my team and me."

"Yin Qi Wenyuan" makes Oracle "come alive" instantly

In order to learn the latest research results and keep our thinking current, whenever Oracle research experts come to Anyang, Liu Yongge and his team members must actively seek advice; they must attend Oracle academic research conferences no matter how tight their time is...

As he learned more and more deeply, Liu Yongge discovered that our nation's oldest writings were obscure and difficult to understand and had few research materials, and were incompatible with real-world technology, resulting in low research efficiency.

How to share Oracle research information, allow all human resources to communicate, and work together to study together? Liu Yongge decided to use his expertise to draw a blueprint for the development of Oracle.

Therefore, a project "Yin Qi Wenyuan" that collects oracle bone inscriptions was quietly launched. This project is to provide the world with all the information about oracle bone inscriptions on the Internet, so that people all over the world can understand, understand and study oracle bone inscriptions. In 2019, the world's first free Oracle database "Yin Qi Wenyuan", which was carefully built by Liu Yongge's team in 8 years, made a stunning debut.

"For example, if you don't know a certain word, you can log in to our Yin Qi Wenyuan. By clicking on the radical, you can find all the information related to the word and input the word into the document." Liu Yongge gave an example.

From the first day when "Yin Qi Wenyuan" was opened, the platform has provided free services to users around the world. It currently has 152 kinds of descriptions and is currently the largest Oracle database. “It not only serves experts in oracle bone inscriptions, archaeology, history, and writing around the world, but also attracts use by lovers of ancient writing, Chinese teachers in primary and secondary schools, and calligraphy enthusiasts.”

Previously, Oracle data was difficult to find, and even Oracle experts could not have all the data. This problem was solved after the completion of the "Yin Qi Wenyuan" project.

Technological innovation empowers deciphering the “civilization code” of oracle bones

Just a while ago, Liu Yongge's laboratory cooperated with Tencent to jointly carry out human-machine collaboration to assist in the research and inheritance of oracle bone inscriptions. This project mainly carries out two aspects of work.

"On the one hand, artificial intelligence technology is used to provide intelligent tools for interpreting oracle bone inscriptions, such as ancient text graphic retrieval tools. These tools will be set up in 'Yinqi Wenyuan'; on the other hand, Tencent's communication platform advantages are used to spread oracle bone inscriptions, such as The WeChat mini-program 'Great Oracle' we are working on will spread the Oracle culture among the public, especially primary and secondary schools, and carry out research activities. It will be developed from the province to the whole country. The mini-program will undergo internal testing at the end of March." Liu Yongge said.

In order to complete the information collection and entry work of the Oracle big data platform, students participate in Oracle informatization research every year starting from their freshman year. With the support of Liu Yongge, the college also launched a special course "Oracle Information Processing".

"Oracle research experts, calligraphy experts, information technology experts... At present, we have established a cross-professional and multi-disciplinary Oracle information processing team. Oracle information teaching and scientific research are at the forefront of the country." Liu Yongge said proudly.

“Currently, we have achieved three technological breakthroughs in cooperation with Tencent. One is to propose a full information digital model of Oracle, the other is to develop the Oracle high-definition shooting technology of illumination transformation matrix, and the third is to establish the technology and technology of Oracle 3D modeling. equipment, the first batch of Oracle 3D modeling experiments will be carried out immediately."

"Do a good job in inheritance and build 'Yin Qi Wenyuan' well, so that Anyang can truly become the world's oracle bone data center; boldly innovate, use artificial intelligence technology to decipher oracle bones, let science and technology empower research, use computers to decipher oracle bone inscriptions, and uncover the behind-the-scenes of oracle bones.' "Code of Civilization', tell the story of oracle bone inscriptions." Liu Yongge said that he wanted to bring the study of oracle bone inscriptions into a new era of intelligence. (Zang Xiaojing)

Source: Dahe.com