Question Answering With RBM-Based Answer-Library Annotations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional recommendation systems in customer service provide only answer libraries without additional useful information or sound advice, making it difficult to effectively assist customer service personnel.
Innovation Solution
A collaborative filtering recommendation method using a restricted Boltzmann machine to determine a set of questions that an answer library can answer and their association relationships, enabling annotation information to be provided to customer service personnel.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a conventional recommendation system is used to provide only answer libraries, then the system complexity is low, but the usefulness and effectiveness for customer service personnel is insufficient
Solution Approach 1:
A restricted Boltzmann machine (RBM) is introduced as an intermediary component between the question and the answer library. The RBM processes the question, determines the answer library, and generates annotation information that explains the association relationships. This intermediary enables the system to provide not only answer libraries but also useful annotations and recommendations, thereby improving reliability without requiring a complete redesign of the entire system architecture.
Solution Approach 2:
The system performs preliminary actions by pre-training the restricted Boltzmann machine with a large corpus of questions and answer libraries before actual use. This pre-training allows the RBM to quickly and accurately determine answer libraries and generate relevant annotations when new questions are submitted, improving the system's effectiveness without adding significant complexity during operation.
2Measurement precision
If a restricted Boltzmann machine is introduced to provide annotation information, then the quality of recommendations improves, but the computational complexity increases
Solution Approach 1:
The system changes parameters by using a restricted Boltzmann machine with specific architectural constraints (binary units, no intra-layer connections) and training techniques. These parameter changes enable the model to capture complex association relationships between questions and answer libraries while keeping the computational complexity manageable through efficient training and inference processes.
Solution Approach 2:
The patent replaces traditional mechanical search methods with a neural network-based restricted Boltzmann machine that uses probabilistic computing. This substitution allows the system to handle complex pattern recognition and association tasks more efficiently than traditional algorithms, improving recommendation quality while controlling computational requirements through the inherent properties of the RBM architecture.
Data Source
AI summary
Embodiments of the present disclosure relate to a question answering method, an electronic device, and a computer program product. The method includes: determining an answer library associated with a question; determining a restricted Boltzmann machine associated with the answer library, wherein the restricted Boltzmann machine is configured to determine a set of questions that the answer library can answer and association relationships between questions in the set of questions and the answer library; and determining, using the restricted Boltzmann machine, annotation information associated with the question and targeted to the answer library. With the technical solution of the present disclosure, it is possible to determine, while determining an answer library associated with the question, annotation information that is targeted to the answer library, and to enable customer service personnel to have a more thorough understanding of the determined answer library and obtain targeted recommendation information.