Overview: Semantic Index transforms CloudOffix into an intelligent knowledge base by enabling search by meaning, not just keywords. Instead of relying on exact word matches, the system understands intent, context, and similarity, allowing users to find the right records and documents—even when the same words are not used. This capability works across CloudOffix records (CRM, Helpdesk, etc.) and external documents (such as PDFs or Word files) stored in connected systems like a DMS (e.g., M-Files), creating a unified and smarter search experience.
Updates:
1. Meaning-Based (Semantic) Search:
- Search results are based on semantic similarity rather than exact keyword matching.
- Related concepts are matched even if the searched word does not explicitly appear.
2. Unified Index for Internal & External Data:
- Semantic Index works on both CloudOffix records (via AI Metadata) and external documents (via External Data Sources) or DMS.
- Users can search across structured records and unstructured files at the same time.
3. Smart Relevance with Meta Models:
- Relational fields (meta data) are included in search relevance using an average importance logic.
- This prevents related data (e.g., company names) from overpowering the actual content meaning.
4. Automatic Background Processing
- Indexing, vector creation, and document preparation run automatically in the background.
- Once configured, no manual intervention is required during daily usage.
Benefits:
• Search by Meaning, Not Keywords: Find relevant results even when different words or languages are used.
• Unified Knowledge Discovery: Search CloudOffix data and external documents from a single AI-powered layer.
• Higher Search Accuracy: Context-aware relevance avoids misleading results caused by unrelated metadata.
• Multilingual Understanding: Search in one language and retrieve results written in another supported language.
How it works?
- Semantic Index works by indexing content from external documents stored in connected systems (e.g., a DMS like M-Files). After configuration, indexing runs automatically in the background.
- Under Semantic Index → External Data Sources, the user adds a new data source and selects the relevant connector (e.g., M-Files).
- Once the configuration is completed, the data source is set to Active.
- From this point on, CloudOffix automatically starts fetching documents from MFiles. No manual action is required.
- Fetched files are processed in the background. Documents are read, cleaned, and split into meaningful chunks so the AI can understand context instead of keywords.
- When processing is completed, documents appear under Indexed Documents. At the same time, CloudOffix records that have Semantic Index enabled are included in the same knowledge layer.
- Users can now search across CloudOffix records, external documents, and DMS content together. Results are matched by meaning, not exact words.