Bext Corporation for an analytical AI agent on May 2, 2025, at the Microsoft AI Co-Innovation Lab KOBE, an AI co-creation lab, with the aim of commercializing innovative technology that combines AI with text mining from the Vext series, centered on VextMiner.
Until now, the role of text mining has been to analyze large amounts of text data such as customer feedback and internal documents to extract business value, but there have been various challenges in obtaining analysis results, such as the barrier of expertise dependent on a few experts, differences in results and interpretations by analysts even for the same data, and the difficulty of performing analysis with high real-time speed. It is the analyst (human) who operates the text mining tool, and it is necessary to control it to obtain the information the analyst needs according to the purpose of the analysis.
Also: NTT DATA launches AI agent cutting marketing work 60%
Therefore, by linking an AI agent with a text mining tool, most of the tool operation will be performed by the AI agent, and when the purpose of the analysis and text data are input to the agent, it will autonomously operate the text mining tool to extract the results, pass them on to another agent to interpret the analysis results, and obtain concrete business improvement measures as output from practical insights. This will evolve text data analysis from “a special task that only a few experts can perform” to “a tool that anyone can use when necessary.” With the support of Microsoft, we developed this technology using Agents and Model Context Protocol (MCP) using the Semantic Kernel framework to put such technology into practical use.
SOURCE: PRTimes
