Relational databases and deep learning models can be seamlessly integrated to enable efficient serving of deep learning inference queries on relational data.
A novel DeepMapping abstraction that leverages deep neural networks to integrate compression and indexing capabilities for efficient storage and retrieval of tabular data.
本文介紹了一個大規模的公開多模態銀行數據集MBD,包含超過150萬個企業客戶的950M筆銀行交易、1B個地理位置事件、500萬條與技術支持的對話以及4種銀行產品的月度購買情況。利用這個數據集,我們提出了兩個實際的業務任務:營銷預測和客戶匹配。我們的實驗結果表明,多模態方法優於單模態方法,為未來的多模態事件序列分析提供了新的視角。
Structured-GraphRAG, a versatile framework, enhances information retrieval across structured datasets by leveraging multiple knowledge graphs to provide more accurate and comprehensive responses to natural language queries.
GraphAr is an efficient storage scheme designed to enhance the capabilities of data lakes for managing Labeled Property Graphs (LPGs). It leverages the strengths of Parquet while introducing specialized techniques to optimize critical graph operations like neighbor retrieval and label filtering.
An adaptive cost model that dynamically optimizes CPU- and I/O-related plan cost parameters at runtime to improve the accuracy of query execution cost estimation and guide the database optimizer towards more optimal query plans.
本文研究了在度量空間中修復不一致資料庫的計算複雜性問題。目標是在最小化原始值與修復值之間的總距離的同時,更新資料庫值以保持一致性。我們考慮了所謂的共現約束,包括鍵約束、包含約束、外鍵約束以及任何對不同標籤(屬性)的單元格數量關係的限制。
데이터베이스 값이 메트릭 공간에 속하는 경우, 동시성 제약을 만족하면서 원래 값과의 거리를 최소화하는 데이터베이스 수리 문제를 효율적으로 해결할 수 있다.
メトリック空間上のデータベースにおいて、一致制約を満たすように最小限の変更で修復する問題を研究した。
The goal is to efficiently compute an optimal repair of an inconsistent database, where database values belong to an underlying metric space, and the repair must satisfy coincidence constraints on the relationships between cell values.