AI-Enabled Search for Storage Systems Using Weighted Metadata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search algorithms are inefficient and inaccurate when dealing with big data, particularly unstructured or semi-structured data, as they rely on exact matching and fail to return relevant results due to the lack of semantically matching text, leading to missed data items.
Innovation Solution
Implementing an AI-enabled search system that uses an AI engine to generate descriptive metadata labels for objects, creating a metadata database with both relational and navigational sections, allowing for contextual or implicit searches based on weighted relationships between objects, enabling the retrieval of relevant data items even in unstructured data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If typical search algorithms use exact matching to search for data, then the search process is simple and fast, but the search accuracy decreases and relevant data items are missed
Solution Approach 1:
The system performs preliminary actions by generating descriptive metadata labels for data items before the search operation. The AI engine analyzes and labels unstructured and semi-structured data in advance, creating a enriched metadata database that enables accurate searching without requiring complex algorithms during the actual search process.
Solution Approach 2:
The patent introduces metadata labels as an intermediary between the search query and the actual data. Instead of directly comparing search terms with raw data, the system uses AI-generated labels that capture the semantic meaning of data items, enabling accurate matching while keeping the search algorithm itself relatively simple.
2Quantity of substance
If storage capacity is increased to accommodate more big data, then the storage capability is improved, but the efficiency of analyzing or searching specific data items decreases
Solution Approach 1:
The system segments the large volume of big data into individually labeled data items. The AI engine processes and labels each data item separately, creating discrete metadata entries that can be efficiently searched. This segmentation transforms the unsearchable mass of big data into structured, queryable units.
Solution Approach 2:
The patent replaces traditional mechanical search methods (keyword matching, indexing) with an AI-based semantic labeling system. Instead of relying on exact text matches, the system uses AI to understand and label the content meaning, enabling efficient searching across large data volumes without the limitations of conventional approaches.
3Measurement precision
If strict matching algorithms are used to ensure precise search results, then the search precision is improved, but the number of relevant results decreases and many data items are missed
Solution Approach 1:
The system changes the parameters used for matching by transitioning from exact text matching to semantic label matching. The AI engine generates labels that capture the essence and meaning of data items, allowing the search to match based on conceptual similarity rather than literal text equality. This parameter change enables both precision and comprehensiveness in search results.
Data Source
AI summary
The present disclosure describes apparatuses and methods for artificial intelligence-enabled search of a storage system. In some aspects, a metadata manager of a storage system receives a label of an object that an AI engine detects in data stored in the storage system. The metadata manager creates, in a relational section of a metadata database, an entry for the detected object with an identifier of the label of the detected object and an address of a node corresponding to the detected object. The metadata manager also creates, in a navigational portion of the metadata database and with the address of the detected object, the node that includes a reference to a relative node of another object and a weight of a relationship between the node and the relative node. By so doing, the metadata database may be searched based on weighted relationships between various nodes, thereby enabling contextual or implicit search of data in the storage system.