Data analysis and integration specialist ADATA Inc. (CEO Lee Dong-hyuk) announced on the 24th that it has recently obtained a patent for machine learning–based preprocessing, a core technology of its "A-rranger" solution.

"A-rranger" is a data preprocessing solution for data analysis. It automates the preprocessing process by applying an intuitive UI/UX, automation of repetitive tasks, and an in-memory DB approach. It was launched in March.

According to ADATA, the registered patent is titled "Machine Learning-Based Data Preprocessing Device and Method." This technology recommends the most efficient preprocessing approach to users through machine learning, enabling them to perform preprocessing tasks without the need for complex script writing when implementing AI systems.

Data analysis consists of four stages—data collection, data preprocessing, data modeling, and interpretation/visualization. Among these, the data preprocessing stage accounts for over 80% of the total process time and is considered the most challenging.

A company representative stated, "This patented technology enhances the efficiency of complex and challenging preprocessing tasks, making data preprocessing easier to perform. The preprocessing process directly impacts data analysis results, which is why it is regarded as an innovative technology."

ADATA also noted that its intuitive UI/UX and no-code approach allow even non-developers to easily carry out data preprocessing tasks.

CEO Lee Dong-hyuk added, "'A-rranger' revolutionizes data preprocessing, reducing the burden of data work and increasing productivity. It will also help accelerate AI system development through integration with our in-development data visualization solution."

Meanwhile, ADATA has demonstrated through trials at public institutions that the labor costs associated with data preprocessing time can be dramatically reduced.

Related Articles

https://news.mt.co.kr/mtview.php?no=2023102413051493685