Query-Based Recommendation System Using ML Classifiers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large corporate entities face challenges in knowledge sharing across geographically dispersed members, making it difficult to identify the right employee to answer specific queries efficiently, leading to unnecessary network bandwidth usage and computing load.
Innovation Solution
A query-based recommendation system using machine learning-trained classifiers represents queries and users as vectors, allowing for the identification of the most likely candidate user to answer a query by determining latent features and generating a likelihood score, thereby reducing unnecessary queries and optimizing resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional knowledge sharing methods are used in large corporate entities, then all members can access the system, but it becomes difficult to identify the right employee to answer specific queries efficiently
Solution Approach 1:
The patent replaces manual browsing and searching methods with a machine learning-based recommendation system. The system uses trained classifiers to automatically analyze query characteristics and recommend the most suitable employee, substituting mechanical human search efforts with an automated intelligent system that efficiently matches queries to experts.
2Reliability
If queries are sent to multiple employees to ensure coverage, then the likelihood of finding an answer increases, but network bandwidth and computing load increase unnecessarily
Solution Approach 1:
The system performs preliminary analysis of the query using trained machine learning classifiers to predict the most likely candidate employee before actually sending the query. This preliminary action filters out unnecessary query transmissions, ensuring that queries are sent only to the most relevant employees, thereby maintaining high answer-finding reliability while minimizing network bandwidth and computing resource consumption.
3Productivity
If a machine learning-based recommendation system is implemented, then query routing efficiency improves, but system complexity and training requirements increase
Solution Approach 1:
The patent extracts the complex machine learning model training and inference processes into a separate, dedicated recommendation system module. This allows the core query routing functionality to remain simple while the complex ML operations are handled by a specialized subsystem, effectively isolating complexity and making the overall system more manageable and maintainable.
Data Source
AI summary
Systems and methods for query-based recommendation systems using machine learning-trained classifiers are provided. A service provider server receives, from a communication device through an application programming interface, a query in an interaction between the server provider server and the communication device. The service provider server generates a vector of first latent features from a set of first visible features associated with the query using a machine learning-trained classifier. The service provider server generates a likelihood scalar value indicating a likelihood of the query is answered by a candidate user in a set of users using a combination of the vector of first latent features and a vector of second latent features. The service provider server provides, to the communication device through the application programming interface, a recommendation message as a response to the query, where the recommendation message includes the likelihood scalar value and an indication of the candidate user.