AnalyticDB:QueryKnowledgeBasesContent

The weight of each SourceCollection.

The filter condition that is used to query data. Specify this parameter in a format that is the same as the WHERE clause. The parameter is an expression that returns a Boolean value of TRUE or FALSE. The condition can be a simple comparison using operators such as equal (=), not equal (<> or !=), greater than (>), less than (<), greater than or equal (>=), or less than or equal (<=). It can also be a more complex expression combining multiple conditions with logical operators (AND, OR, NOT), or use keywords such as IN, BETWEEN, and LIKE.

• For the syntax, see https://www.postgresqltutorial.com/postgresql-tutorial/postgresql-where/.

Whether to enable knowledge graph enhancement. Default value: false.

Returns the top number of entities and relationship edges. Default value: 60.

The dual-path retrieval algorithm. This parameter is empty by default, which specifies that scores of vector retrieval and full-text retrieval are directly compared and sorted together.

Valid values:

• RRF: The reciprocal rank fusion (RRF) algorithm uses a constant k to control the fusion effect. For more information, see the description of the HybridSearchArgs parameter.
• Weight: This algorithm uses the alpha parameter to specify the proportion of the vector search score and the full-text search score and then sorts by weight. For more information, see the description of the HybridSearchArgs parameter.
• Cascaded: This algorithm performs first full-text retrieval and then vector retrieval.

The parameters of the dual-path retrieval algorithm. RRF and Weight are supported at this time:

• RRF: Specifies the smoothing constant k in the formula to calculate the score: 1/(k + rank_i). The k constant must be a positive integer greater than 1. The format:

{ 
   "RRF": {
    "k": 60
   }
}

• Weight: The score is computed as alpha * vector_score + (1 - alpha) * text_score. The parameter alpha controls the weighting between vector search and full-text search scores, with a valid range of [0, 1]. 0 specifies only full-text search score. 1 specifies only vector search score.

{ 
   "Weight": {
    "alpha": 0.5
   }
}

The method that is used to create vector indexes. Valid values:

• l2: Euclidean distance.
• ip: Inner product distance.
• cosine: Cosine similarity.

The retrieval window. If you specify this parameter, the context of the retrieved result is added in the output. Format: List<A, B>. Valid values: -10<=A<=0 and 0<=B<=10.

• We recommend that you specify this parameter if the source document is segmented into large numbers of pieces, which may result in loss of contextual information during retrieval.

• Perform re-ranking before windowing.

The rerank factor. If you specify this parameter, the vector retrieval results are reranked once more. Valid values: 1<RerankFactor<=5.

• If the document is segmented into sparse parts, reranking is inefficient.

• We recommend that the number of reranked results (the ceiling of TopK × RerankFactor) not exceed 50.

The number of top results.

Specifies whether to use full-text retrieval (dual-path retrieval). The default value is false, which means only vector retrieval is used.

The fields by which to sort the results. This parameter is empty by default.

The field must be either a metadata field or a default field in the table (e.g., id). Supported formats include:

Single field, such as chunk_id. Multiple fields that are separated by commas (,), such as block_id,chunk_id. Descending order is supported, such as block_id DESC,chunk_id DESC.

Offset for pagination.