The qKnow Agent Building Platform Professional Edition v3.1.2 is now live, featuring significant optimizations to knowledge file data synchronization.
This release introduces one-click import capabilities for third-party storage solutions, including HDFS, OSS, and FTP, further streamlining the data ingestion process for both knowledge bases and knowledge graphs.
By enhancing data access efficiency, this update ensures that files scattered across various storage systems can be seamlessly integrated into qKnow’s processing workflows.
The Value of Third-Party Storage Synchronization
In real-world enterprise scenarios, knowledge files are rarely centralized.
Large-scale business documents often reside in distributed file systems, cloud-based materials are stored in object storage platforms, and historical files are typically managed via FTP.
Traditionally, integrating these into a knowledge base required a cumbersome process: downloading from third-party storage, organizing locally, and manually uploading for parsing. This approach increased manual labor and delayed data updates.
Version 3.1.2 eliminates this friction by establishing direct connections between external storage and the knowledge base, significantly reducing operational complexity.
Core Capability Upgrades
Knowledge Base Third-Party Storage Sync:
The new release introduces an OSS synchronization feature for knowledge documents. Users can directly import files from third-party storage into the knowledge base via the document management page.
The process involves two simple steps:
- Configure Data Connection: Select the storage type (HDFS for large-scale distributed storage, FTP for batch file reading, or Alibaba Cloud OSS for cloud data) and enter the connection details.
- Select Files for Sync: Once connected, the system reads the external file list, allowing users to select specific files for direct import, eliminating redundant data transfer.
External File Sync for Knowledge Graph Unstructured Extraction:
Beyond the knowledge base, v3.1.2 enhances the unstructured extraction workflow for knowledge graphs.
A new OSS sync button has been added to the extraction module. Users can connect to third-party storage and select files for knowledge extraction using the same streamlined process, laying a solid data foundation for knowledge graph construction.
Optimization & Bug Fixes
Alongside new synchronization features, this release includes several experience enhancements:
- Fixed an issue where the model dropdown in relationship configuration displayed incomplete data, improving selection usability.
- Enhanced knowledge graph model management by allowing concepts with identical names to be created across different graph models, increasing design flexibility.
- Improved the display of model classification titles on the large model configuration page for clearer structural organization.
- Increased the stability of unstructured extraction tasks; isolated task failures no longer interrupt the execution of other tasks.
- Optimized blank paragraph handling during extraction, skipping empty segments to reduce invalid processing and improve overall efficiency.
Release Value
qKnow Professional Edition v3.1.2 focuses on strengthening data ingestion for knowledge bases and graphs. By supporting synchronization with HDFS, OSS, and FTP, it enables seamless integration of distributed knowledge files into the agent building process.
This upgrade delivers value across three main areas: reducing knowledge data ingestion costs, completing the knowledge construction data pipeline, and enhancing overall platform stability.
Moving forward, qKnow will continue to optimize capabilities around knowledge enhancement, agent building, and enterprise-level knowledge management, driving greater efficiency across the entire lifecycle from data ingestion and comprehension to practical application.
Top comments (0)