VERSES AI Inc. has filed a provisional patent application for a new method for Predictive Querying on Vector Graph Document databases. This method aims to address the limitations of prior arts by allowing probabilistic querying on vector graph document databases.
Predictive queries are designed to provide users with additional information that is predicted to be of interest to them, based on the context of the query. VERSES’ novel Predictive Query method utilizes Hyperspatial Modeling Language (HSML) and an inference algorithm to generate probabilistic and contextualized results.
This method is significant because it is the first querying method that enables probabilistic querying on vector graph document databases. It allows the engine to generate predictions about the information being searched for by the user, based on comparative, relationship, and similarity information.
HSML is a modeling language that represents the relationships between entities in a knowledge graph. An HSML vector graph document database structures data as an HSML knowledge graph, enabling complex queries that involve entity comparison, cause-effect relations, and entity similarity.
Compared to other classes of databases, such as vector search and graph databases, the new method for Predictive Querying offers more effective modeling, managing, and querying of data tailored to new modalities in artificial intelligence.
This development signifies VERSES’ leadership in the AI landscape. The company’s CTO, Jason Fox, expressed pride in the team and the tools they are building. With this method, VERSES can return the most probable and relevant matches to a user’s search query, based on their implicit goal.
VERSES AI Inc. is a cognitive computing company specializing in next-generation Artificial Intelligence. Their flagship offering, GIA™, is an Intelligent Agent powered by the KOSM™ network operating system, which enables distributed intelligence. VERSES aims to create a smarter world that leverages human potential through nature-inspired innovations.
The post A Method for Predictive Querying on Vector Graph Document Databases appeared first on TS2 SPACE.
