{"id":624,"date":"2026-07-04T14:55:17","date_gmt":"2026-07-04T05:55:17","guid":{"rendered":"https:\/\/conceptminer.ai\/?page_id=624"},"modified":"2026-07-04T15:15:34","modified_gmt":"2026-07-04T06:15:34","slug":"624-2","status":"publish","type":"page","link":"https:\/\/conceptminer.ai\/?page_id=624&lang=en","title":{"rendered":"ConceptMiner Concept Index"},"content":{"rendered":"\n
\n\n
\n
\n
\n
ConceptMiner Concept Index<\/div>\n

Search and classify free-text data by \u201cconcept\u201d inside your existing PostgreSQL database.<\/h1>\n

\n Without introducing a new vector database or graph database,\n Concept Index enables you to extract semantically similar groups\n from VoC data, inquiry histories, free-text survey responses,\n sales notes, project summaries, and other business records.\n <\/p>\n

\n View Limited Pre-order Offer<\/a>\n See How It Works<\/a>\n <\/div>\n

\n Concept Index is scheduled for release in August 2026. The initial supported database is PostgreSQL.\n <\/p>\n <\/div>\n\n

\n
\n
\n
DB<\/div>\n
\n Existing RDB<\/strong>\n Business data stored in PostgreSQL<\/span>\n <\/div>\n <\/div>\n
\n
TXT<\/div>\n
\n Analyze Text Columns<\/strong>\n Free text, inquiries, and business notes<\/span>\n <\/div>\n <\/div>\n
\n
AI<\/div>\n
\n Generate Concept Nodes<\/strong>\n Structure semantic groups using GNG + MST<\/span>\n <\/div>\n <\/div>\n
\n
IDX<\/div>\n
\n Add Concept Index<\/strong>\n Assign node IDs and neighborhood information to each record<\/span>\n <\/div>\n <\/div>\n
\n
SQL<\/div>\n
\n Extract in RDB<\/strong>\n Search similar groups and concept-based segments<\/span>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/section>\n\n\n
\n
\n
\n
Problem<\/div>\n

Free-text data is difficult to use, even when it is already stored in an RDB.<\/h2>\n

\n Companies accumulate large amounts of valuable text data, such as customer feedback,\n inquiry histories, sales notes, and survey responses.\n However, conventional RDBs make it difficult to handle \u201csemantic similarity\u201d\n that cannot be captured by keyword search or fixed categories.\n <\/p>\n <\/div>\n\n

\n
\n The Limit of Keyword Search<\/strong>\n Records with the same meaning may not be found when they use different wording.<\/span>\n <\/div>\n
\n The Limit of Manual Classification<\/strong>\n Manual classification takes time and depends heavily on individual judgment.<\/span>\n <\/div>\n
\n The Burden of Introducing a New Database<\/strong>\n Adding a vector database or graph database can make operations more complex.<\/span>\n <\/div>\n <\/div>\n <\/div>\n <\/section>\n\n\n
\n
\n
\n
Solution<\/div>\n

Concept Index assigns \u201cconcept nodes\u201d to each record.<\/h2>\n

\n Concept Index analyzes business data that includes text columns\n and organizes records with similar meanings into concept nodes.\n As a result, each record gains information not only as an ID or category,\n but also as a member of a specific concept group.\n <\/p>\n <\/div>\n\n

\n
\n Existing RDB<\/b>\n Business data stored in PostgreSQL or similar databases<\/small>\n <\/div>\n
\n Embedding<\/b>\n Represent text columns as semantic vectors<\/small>\n <\/div>\n
\n GNG + MST<\/b>\n Learn concept structures from semantically similar data groups<\/small>\n <\/div>\n
\n Node Assignment<\/b>\n Add concept node IDs to each record<\/small>\n <\/div>\n
\n RDB Search<\/b>\n Extract similar groups and neighboring nodes using SQL<\/small>\n <\/div>\n <\/div>\n <\/div>\n <\/section>\n\n\n
\n
\n
\n
Use Cases<\/div>\n

Designed for these types of business data.<\/h2>\n

\n Concept Index is suitable when you want to extract groups of data\n that are semantically similar but difficult to find through keywords alone.\n <\/p>\n <\/div>\n\n

\n
\n
\ud83d\udcac<\/div>\n

VoC Analysis<\/h3>\n

Extract customer groups with similar complaints, requests, or expectations, and use them for product and service improvement.<\/p>\n <\/div>\n

\n
\ud83d\udce9<\/div>\n

Inquiry Histories<\/h3>\n

Identify similar consultation topics and recurring trouble patterns to improve response quality.<\/p>\n <\/div>\n

\n
\ud83d\udcdd<\/div>\n

Free-text Survey Responses<\/h3>\n

Organize respondents\u2019 interests, concerns, and expectations as meaningful semantic groups.<\/p>\n <\/div>\n

\n
\ud83e\udd1d<\/div>\n

Sales Notes<\/h3>\n

Extract similar sales patterns and deal types to support the knowledgeization of sales activities.<\/p>\n <\/div>\n

\n
\ud83d\udcc2<\/div>\n

Project Summaries<\/h3>\n

Refer to past similar projects by concept and use them as material for proposals and decision-making.<\/p>\n <\/div>\n

\n
\ud83e\udded<\/div>\n

Business Data Exploration<\/h3>\n

Discover latent patterns in operational data that are difficult to see through fixed categories.<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/section>\n\n\n

\n
\n
\n
Important<\/div>\n

This is not RAG. Concept Index is designed to find similar data groups, not individual answers.<\/h2>\n

\n Concept Index is not a search engine for returning pinpoint documents in response to a question.\n Its purpose is to extract conceptually similar groups and segments from records that include free-text data.\n If you want to build a question-answering system or knowledge base, please use\n ThinkNavi Knowledge Base Builder instead.\n <\/p>\n <\/div>\n <\/div>\n <\/section>\n\n\n

\n
\n
\n
Difference<\/div>\n

From Top-K search to concept-based segment extraction.<\/h2>\n

\n Concept Index is not designed to search for the nearest document one by one.\n Instead, it uses pre-built concept nodes and neighborhood relationships\n to handle business-meaningful groups of data.\n <\/p>\n <\/div>\n\n \n \n \n \n \n \n \n \n
Comparison Item<\/th>\n General Vector Search<\/th>\n Concept Index<\/th>\n <\/tr>\n <\/thead>\n
Main Purpose<\/td>\n Find documents close to a query<\/td>\n Extract semantically similar data groups<\/span><\/td>\n <\/tr>\n
Data Infrastructure<\/td>\n Often requires an additional vector database or dedicated infrastructure<\/td>\n Uses existing RDB infrastructure<\/span><\/td>\n <\/tr>\n
Search Unit<\/td>\n Individual documents or individual records<\/td>\n Concept nodes, neighboring nodes, and segments<\/span><\/td>\n <\/tr>\n
Suitable Uses<\/td>\n RAG, FAQ, document search<\/td>\n VoC analysis, survey classification, project group extraction<\/span><\/td>\n <\/tr>\n <\/tbody>\n <\/table>\n <\/div>\n <\/section>\n\n\n
\n
\n
\n
Delivery<\/div>\n

Connector-based delivery for easy integration into existing environments.<\/h2>\n

\n Concept Index is introduced by adding a concept index\n to the existing RDB where your business data is stored.\n Rather than building a large new AI infrastructure,\n it extends the way you can use your current database.\n <\/p>\n <\/div>\n\n

\n
\n
\ud83d\udda5\ufe0f<\/div>\n

Windows \/ Linux Support<\/h3>\n

Provided as an executable file to reduce the burden of setting up a Python environment.<\/p>\n <\/div>\n

\n
\ud83d\udd11<\/div>\n

Set API Key and DB Connection Information<\/h3>\n

Specify the target table or View and run the concept index construction process.<\/p>\n <\/div>\n

\n
\ud83e\udde9<\/div>\n

Write Results Back to the RDB<\/h3>\n

Make concept node IDs, neighborhood relationships, and classification results available inside the RDB.<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/section>\n\n\n

\n
\n
\n
Roadmap<\/div>\n

In the future, extract concept-based segments using natural language.<\/h2>\n

\n As a future extension, we are considering Concept Query,\n a feature that allows users to specify complex segment conditions in natural language.\n <\/p>\n <\/div>\n\n

\n
\n

Customers with high profitability but also high churn risk<\/h3>\n

Concept-based customer extraction that combines quantitative data with free-text data.<\/p>\n <\/div>\n

\n

Groups with low price sensitivity and strong premium orientation<\/h3>\n

Discovery of purchasing tendencies that are difficult to see through simple attribute classification.<\/p>\n <\/div>\n

\n

Segments with many inquiries but also high satisfaction<\/h3>\n

Understand not only problem frequency, but also improvement potential and relationship strength.<\/p>\n <\/div>\n <\/div>\n\n

\n \u203b Concept Query is currently under consideration as a future extension.\n <\/p>\n <\/div>\n <\/section>\n\n\n

\n
\n
\n
Pre-order<\/div>\n

Limited Pre-order Offer for Concept Index<\/h2>\n

\n ConceptMiner Concept Index, scheduled for release in August 2026,\n is available at a limited first-year pre-order price.\n <\/p>\n <\/div>\n\n

\n
\n

Standard Annual License<\/h3>\n
$2,000.00 tax excluded \/ year<\/small><\/div>\n