Knowledge Bases support two primary querying approaches: semantic search and metadata filtering, each of which offers different filtering capabilities, including filtering by the relevance score to ensure only data most relevant to the query is returned.

  • Semantic Search

    Semantic search enables users to query Knowledge Bases using natural language. When searching semantically, you reference the content column in your SQL statement. MindsDB will interpret the input as a semantic query and use vector-based similarity to find relevant results.

    SELECT * FROM my_kb
WHERE content = 'what document types store reviews?';

  • Metadata Filtering

    It allows users to query Knowledge Bases based on the available metadata fields. These fields can be used in the WHERE clause of a SQL statement.

    SELECT * FROM my_kb
WHERE document_type = ‘cover letter’
AND document_author = 'bot';

  • Relevance Filtering

    Every semantic search result is assigned a relevance score, which indicates how closely a given entry matches your query. You can filter results by this score to ensure only the most relevant entries are returned.

See more examples here.

​
SELECT FROM KB Syntax

Knowledge bases provide an abstraction that enables users to see the stored data.

Note that here a sample knowledge base created and inserted into in the previous Example sections is searched.

SELECT *
FROM my_kb;

Here is the sample output:

+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+
| id  | chunk_id             | chunk_content           | metadata                                                                                                                                                                                            | distance           | relevance          |
+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+
| A1B | A1B_notes:1of1:0to20 | Request color: black    | {"chunk_index":0,"content_column":"notes","end_char":20,"original_doc_id":"A1B_notes","original_row_id":"A1B","product":"Wireless Mouse","source":"TextChunkingPreprocessor","start_char":0}        | 0.5743341242061104 | 0.5093188026135379 |
| Q7P | Q7P_notes:1of1:0to22 | Prefer aluminum finish  | {"chunk_index":0,"content_column":"notes","end_char":22,"original_doc_id":"Q7P_notes","original_row_id":"Q7P","product":"Aluminum Laptop Stand","source":"TextChunkingPreprocessor","start_char":0} | 0.7744703514692067 | 0.2502580835880018 |
| 3XZ | 3XZ_notes:1of1:0to19 | Gift wrap requested     | {"chunk_index":0,"content_column":"notes","end_char":19,"original_doc_id":"3XZ_notes","original_row_id":"3XZ","product":"Bluetooth Speaker","source":"TextChunkingPreprocessor","start_char":0}     | 0.8010851611432231 | 0.2500003885558766 |
+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+

The following columns are stored in the knowledge base.

  • id It stores values from the column defined in the id_column parameter when creating the knowledge base. These are the source data IDs.

  • chunk_id Knowledge bases chunk the inserted data in order to fit the defined chunk size. If the chunking is performed, the following chunk ID format is used: <id>:<chunk_number>of<total_chunks>:<start_char_number>to<end_char_number>.

  • chunk_content It stores values from the column(s) defined in the content_columns parameter when creating the knowledge base.

  • metadata It stores the general metadata and the metadata defined in the metadata_columns parameter when creating the knowledge base.

  • distance It stores the calculated distance between the chunk’s content and the search phrase.

  • relevance It stores the calculated relevance of the chunk as compared to the search phrase. Its values are between 0 and 1.

Note that the calculation method of relevance differs as follows:

  • When the ranking model is provided, the default relevance is equal or greater than 0, unless defined otherwise in the WHERE clause.
  • When the reranking model is not provided and the relevance is not defined in the query, then no relevance filtering is applied and the output includes all rows matched based on the similarity and metadata search.
  • When the reranking model is not provided but the relevance is defined in the query, then the relevance is calculated based on the distance column (1/(1+ distance)) and the relevance value is compared with this relevance value to filter the output.

Users can query a knowledge base using semantic search by providing the search phrase (called content) to be searched for.

SELECT *
FROM my_kb
WHERE content = 'color'

Here is the output:

+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+
| id  | chunk_id             | chunk_content           | metadata                                                                                                                                                                                            | distance           | relevance          |
+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+
| A1B | A1B_notes:1of1:0to20 | Request color: black    | {"chunk_index":0,"content_column":"notes","end_char":20,"original_doc_id":"A1B_notes","original_row_id":"A1B","product":"Wireless Mouse","source":"TextChunkingPreprocessor","start_char":0}        | 0.5743341242061104 | 0.5093188026135379 |
| Q7P | Q7P_notes:1of1:0to22 | Prefer aluminum finish  | {"chunk_index":0,"content_column":"notes","end_char":22,"original_doc_id":"Q7P_notes","original_row_id":"Q7P","product":"Aluminum Laptop Stand","source":"TextChunkingPreprocessor","start_char":0} | 0.7744703514692067 | 0.2502580835880018 |
| 3XZ | 3XZ_notes:1of1:0to19 | Gift wrap requested     | {"chunk_index":0,"content_column":"notes","end_char":19,"original_doc_id":"3XZ_notes","original_row_id":"3XZ","product":"Bluetooth Speaker","source":"TextChunkingPreprocessor","start_char":0}     | 0.8010851611432231 | 0.2500003885558766 |
+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+

When querying a knowledge base, the default values include the following:

  • relevance
    If not provided, its default value is equal to or greater than 0, ensuring there is no filtering of rows based on their relevance.

  • LIMIT
    If not provided, its default value is 10, returning a maximum of 10 rows.

Note that when specifying both relevance and LIMIT as follows:

SELECT *
FROM my_kb
WHERE content = 'color'
AND relevance >= 0.5
LIMIT 20;

The query extracts 20 rows (as defined in the LIMIT clause) that match the defined content. Next, these set of rows is filtered out to match the defined relevance.

Users can limit the relevance in order to get only the most relevant results.

SELECT *
FROM my_kb
WHERE content = 'color'
AND relevance >= 0.5;

Here is the output:

+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+
| id  | chunk_id             | chunk_content          | metadata                                                                                                                                                                                     | distance           | relevance          |
+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+
| A1B | A1B_notes:1of1:0to20 | Request color: black   | {"chunk_index":0,"content_column":"notes","end_char":20,"original_doc_id":"A1B_notes","original_row_id":"A1B","product":"Wireless Mouse","source":"TextChunkingPreprocessor","start_char":0} | 0.5743341242061104 | 0.5103766499957533 |
+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+

By providing the relevance filter, the output is limited to only data with relevance score of the provided value. The available values of relevance are between 0 and 1, and its default value covers all available relevance values ensuring no filtering based on the relevance score.

Users can limit the number of rows returned.

SELECT *
FROM my_kb
WHERE content = 'color'
LIMIT 2;

Here is the output:

+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+
| id  | chunk_id             | chunk_content           | metadata                                                                                                                                                                                            | distance           | relevance          |
+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+
| A1B | A1B_notes:1of1:0to20 | Request color: black    | {"chunk_index":0,"content_column":"notes","end_char":20,"original_doc_id":"A1B_notes","original_row_id":"A1B","product":"Wireless Mouse","source":"TextChunkingPreprocessor","start_char":0}        | 0.5743341242061104 | 0.5093188026135379 |
| Q7P | Q7P_notes:1of1:0to22 | Prefer aluminum finish  | {"chunk_index":0,"content_column":"notes","end_char":22,"original_doc_id":"Q7P_notes","original_row_id":"Q7P","product":"Aluminum Laptop Stand","source":"TextChunkingPreprocessor","start_char":0} | 0.7744703514692067 | 0.2502580835880018 |
+-----+----------------------+-------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+--------------------+

​
Metadata Filtering

Besides semantic search features, knowledge bases enable users to filter the result set by the defined metadata.

SELECT *
FROM my_kb
WHERE product = 'Wireless Mouse';

Here is the output:

+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+-----------+----------+
| id  | chunk_id             | chunk_content          | metadata                                                                                                                                                                                     | relevance | distance |
+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+-----------+----------+
| A1B | A1B_notes:1of1:0to20 | Request color: black   | {"chunk_index":0,"content_column":"notes","end_char":20,"original_doc_id":"A1B_notes","original_row_id":"A1B","product":"Wireless Mouse","source":"TextChunkingPreprocessor","start_char":0} | [NULL]    | [NULL]   |
+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+-----------+----------+

Note that when searching by metadata alone, the relevance column values are not calculated.

Users can do both, filter by metadata and search by content.

SELECT *
FROM my_kb
WHERE product = 'Wireless Mouse'
AND content = 'color'
AND relevance >= 0.5;

Here is the output:

+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+-------------------+
| id  | chunk_id             | chunk_content          | metadata                                                                                                                                                                                     | distance           | relevance         |
+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+-------------------+
| A1B | A1B_notes:1of1:0to20 | Request color: black   | {"chunk_index":0,"content_column":"notes","end_char":20,"original_doc_id":"A1B_notes","original_row_id":"A1B","product":"Wireless Mouse","source":"TextChunkingPreprocessor","start_char":0} | 0.5743341242061104 | 0.504396172197583 |
+-----+----------------------+------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------------+-------------------+

​
JOIN Syntax

Knowledge bases can be used in the standard SQL JOIN statements.

SELECT t.order_id, t.product, t.notes, kb.chunk_content, kb.relevance
FROM local_postgres.orders AS t
JOIN my_kb AS kb
ON t.order_id = kb.id
WHERE t.order_id = 'A1B'
AND kb.content = 'color'
AND kb.product = 'Wireless Mouse';

Here is the output:

+----------+------------------+------------------------+------------------------+--------------------+
| order_id | product          | notes                  | chunk_content          | relevance          |
+----------+------------------+------------------------+------------------------+--------------------+
| A1B      | Wireless Mouse   | Request color: black   | Request color: black   | 0.5106591666649376 |
+----------+------------------+------------------------+------------------------+--------------------+

​
Examples

We have a knowledge base that stores data about movies.

+----------+-----------------------------------+-------------------------------------------------------------------------+
| id       | content                           | metadata                                                                |
+----------+-----------------------------------+-------------------------------------------------------------------------+
| movie_id | "A bank security expert plots..." | {"genre":"Crime","rating":6.3,"expanded_genres":"Comedy, Crime, Drama"} |
+----------+-----------------------------------+-------------------------------------------------------------------------+

It uses the movie_id column to uniquely identify each entry. The content column stores the description of the movie, and the metadata includes genre, rating, and expanded_genre columns.

Let’s see the query examples.

  • Selecting high-rated action movies with heist themes and no romance.

    SELECT * FROM movies_kb 
WHERE content LIKE 'heist bank robbery space alien planet'
AND genre != 'Romance' 
AND expanded_genres NOT LIKE '%Romance%'
AND rating > 7.0;

    This query includes a semantic search filtering condition - content LIKE 'heist bank robbery space alien planet' - and multiple metadata filtering conditions - genre != 'Romance' AND expanded_genres NOT LIKE '%Romance%' AND rating > 7.0.

  • Selecting action-comedies with car chase scenes.

    SELECT * FROM movies_kb 
WHERE content LIKE 'car chase driving speed race'
AND expanded_genres LIKE '%Action%'
AND expanded_genres LIKE '%Comedy%'
AND rating > 6.5;

    This query includes a semantic search filtering condition - content LIKE 'car chase driving speed race' - and multiple metadata filtering conditions - expanded_genres LIKE '%Action%' AND expanded_genres LIKE '%Comedy%' AND rating > 6.5.

  • Selecting historical dramas without war themes.

    SELECT * FROM movies_kb 
WHERE content LIKE 'historical period past century era'
AND content NOT LIKE 'war battle soldier military'
AND content NOT LIKE 'fight combat weapon'
AND expanded_genres LIKE '%Drama%'
AND rating > 3.5;

    This query includes multiple semantic search filtering conditions - content LIKE 'historical period past century era' AND content NOT LIKE 'war battle soldier military' AND content NOT LIKE 'fight combat weapon' - and multiple metadata filtering conditions - expanded_genres LIKE '%Drama%' AND rating > 3.5.

  • Selecting multi-genre movies with different ratings.

    SELECT * FROM movies_kb 
WHERE (content LIKE 'detective mystery investigation' AND (genre = 'Mystery' OR expanded_genres LIKE '%Thriller%'))
OR (content LIKE 'romance love relationship' AND (genre = 'Romance' OR expanded_genres LIKE '%Romance%'))
AND rating > 7.0;

    This query includes nested semantic search filtering conditions - (content LIKE 'detective mystery investigation' AND (genre = 'Mystery' OR expanded_genres LIKE '%Thriller%')) - and a metadata filtering condition - rating > 7.0.

  • Selecting adventure movies excluding some genres.

    SELECT * FROM movies_kb 
WHERE content LIKE 'adventure journey quest treasure'
AND genre NOT IN ('Horror', 'Romance', 'Family')
AND rating > 6.5;

    This query includes multiple semantic search filtering conditions - content LIKE 'adventure journey quest treasure' - and multiple metadata filtering conditions - genre NOT IN ('Horror', 'Romance', 'Family') AND rating > 6.5.

  • Selecting comedy movies in specific rating range.

    SELECT * FROM movies_kb 
WHERE content LIKE 'comedy funny humor laugh'
AND rating BETWEEN 7.0 AND 9.0
AND expanded_genres LIKE '%Comedy%';

    This query includes multiple semantic search filtering conditions - content LIKE 'comedy funny humor laugh' - and multiple metadata filtering conditions - rating BETWEEN 7.0 AND 9.0 AND expanded_genres LIKE '%Comedy%'.

  • Selecting different thriller subgenres.

    SELECT * FROM movies_kb 
WHERE content LIKE 'detective investigation mystery' AND rating > 7.0
UNION
SELECT * FROM movies_kb 
WHERE content LIKE 'heist robbery theft steal' AND rating > 7.0
UNION
SELECT * FROM movies_kb 
WHERE content LIKE 'spy secret agent undercover' AND rating > 7.0;

    This query combines the results of three queries using the UNION operator.