Service Request Retrieval via Hash and Semantic Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large-scale information service providers, technical support teams face inefficiencies in processing service requests due to varying levels of experience among engineers, with junior engineers being less efficient than seniors, and there is a need for an intelligent and automatic solution recommendation system to improve work efficiency and customer satisfaction.
Innovation Solution
A method is provided that converts service request classification into a similarity learning problem, using a two-step similarity modeling approach to determine hash correlations and semantic correlations between service requests, thereby recommending solved service requests with high similarity scores to supporting engineers, improving retrieval efficiency by reducing computing overhead and time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If junior engineers process service requests manually without assistance, then they can handle requests independently, but their work efficiency is lower compared to senior engineers
Solution Approach 1:
The system enables self-service by automatically recommending solutions to junior engineers based on historical service requests and similarity analysis, allowing them to resolve issues independently without requiring senior engineer intervention or extensive manual research
Solution Approach 2:
An intelligent recommendation system acts as an intermediary between historical service requests and current engineers, automatically analyzing similarities and retrieving relevant past cases to assist engineers in resolving new service requests efficiently
2Measurement precision
If all historical service requests are retrieved and analyzed for every new service request, then comprehensive similarity assessment is achieved, but computing overhead and retrieval time increase significantly
Solution Approach 1:
The retrieval process is segmented into multiple stages: initial filtering based on service request categories, then similarity assessment only on filtered subsets, and finally detailed analysis on top candidates. This segmentation reduces the number of requests requiring full analysis while maintaining assessment accuracy
Solution Approach 2:
Preliminary filtering and categorization of historical service requests are performed before detailed similarity analysis, preparing the data in advance so that when a new service request arrives, only relevant pre-filtered requests need to be analyzed in detail, reducing computation time
Data Source
AI summary
Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for retrieving service requests. The method includes determining a hash value of a service request based on the service request. The method further includes determining request pairs based on correlations between the hash value of the service request and a plurality of hash values of a plurality of historical service requests. The method further includes determining a semantic correlation between the service request and the historical service request in each of the plurality of request pairs. The method further includes determining, based on the determined semantic correlation between each request pair, a probability indicating that the service request and the historical service request in the request pair use the same solution. The method further includes determining a retrieved historical service request based on the probability.