Bridging Natural Language with Data Programming for Combinatorial Problems
For combining the common sense and the generative power of LLM with a symbolic engine able to solve computational intensive problems in a precise and predict...
For combining the common sense and the generative power of LLM with a symbolic engine able to solve computational intensive problems in a precise and predict...
a method to efficiently synthesise queries for extracting information from an in-memory hypergraph Knowledge Base. It uses positive and negative samples to g...
An in-memory hyper-graph encoding for Datalog-based programming systems. It defines data structures for direct access to predicates and their arguments, enab...
Modeling complex processes may involve representing and evaluating long chains of time-dependent data values that depend on one another, for example, as caus...
Business processes are encoded in software through phases like analysis, design, implementation, and quality checking. To reduce costs and delays, systems sh...
E-Stores loose sells due to the negative biases of consumers. While salespeople give proper reasons to change consumers misbelieves, it is problematic to add...
Inductive Logic Programming method for synthesizing logic queries to retrieve information from graph databases.
Return items with similar characteristics out of user’s e-commerce search, though the searched product is not known in the merchant’s catalog.
Neural network architecture for smoothing context switching on conversations between human and automatic agents.
This approach provides useful information that could be managed by an conversational engine for corroborating search results with meaningful answers.
Automatic system for maximizing user pereferences’ gathering for optimizing eCommerce product offering