Coreline Soft Launches AI Data Refinement Platform Applying 37 Million Big Data Records

Coreline Soft announced on the 7th that it has launched an AI training data refinement platform called ‘CORE:ALAP (Coreline: AI Labeling & Analysis Platform)’. Based on its data processing and analysis technologies, Coreline Soft released CORE:ALAP (Core A-Lab). Through this, the company plans to strengthen its AI technology and solidify its leading position in the market.

CORE:ALAP emphasizes the crucial ‘Labeling’ technology in the data collection and preprocessing stages, aiming to enhance AI training data refinement capabilities. Data labeling refers to the process of analyzing large volumes of data for AI training, assigning correct answers or meaningful information to each data point, and standardizing it. Especially in the medical field, labeled data is essential to ensure patient safety and medical reliability.

CORE:ALAP differentiates itself by having accumulated a high-quality labeled dataset over a long period and having commercialized products in medical settings. Coreline Soft has supplied various AI software (SW) applied in the medical field, accumulating approximately 37 million datasets used to implement AI engines. Among the training data, 80% consists of proprietary datasets exclusively held by Coreline Soft.

AI training data applied in the medical field is difficult to secure and involves complex source data, so not only the quantity but also the quality of data is considered a critical indicator of product performance. Coreline Soft has elevated AI accuracy and reliability in this field to a world-class level through collaboration with top domestic and international medical professionals and its self-developed 3D work tools optimized for labeling.

To secure high-quality data, Coreline Soft automated the labeling process based on its self-developed framework. By applying cloud technology to the framework, it successfully implemented multiple AI software based on large-scale video data through efficient management of AI training data, and its product performance and technological capabilities were recognized through KLUCAS (Korea Lung Cancer Screening Project).

Lee Jae-yeon, Chief Technology Officer (CTO) of Coreline Soft, said, “Automating labeling to quickly secure high-quality AI training data is Coreline Soft’s core technological competitiveness, and we launched CORE:ALAP to further advance this.” He added, “Based on our technological capabilities, we will enhance AI performance and integrate the entire value chain.”

He continued, “Due to the nature of source data, collecting and refining AI training data applied in the medical field is more challenging than in other industries, so expansion into other industries will also be feasible.” He added, “Based on domestic and global supply references, we will leap forward as a leading company in the video data-based AI field.”