Expert Recommender Engine Using Topic Clusters for Service Requests

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional information processing systems rely on inefficient manual processes to analyze unstructured text data for service events, leading to incomplete expert selection in handling service requests, as they often use static codes and require manual intervention, neglecting the value of unstructured data.

Innovation Solution

Implementing an expert recommender engine that processes unstructured service event data to identify topic clusters, utilizing collaborative filtering with structured data, social media data, and customer satisfaction data to automatically recommend subject matter experts, eliminating the need for manual rule customization and database updates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual processing of unstructured text data is used, then service personnel can review and sample the data, but the process becomes tedious and time-consuming

Engineering Contradiction:

Improvedata processing reliabilityVSAvoidtime consumption

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical processing of unstructured text data with automated machine learning-based natural language processing. The system automatically analyzes service request summaries, identifies topics, and clusters them without human intervention, eliminating the tedious manual screening process while maintaining processing reliability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs self-service by automatically processing and analyzing unstructured text data through embedded machine learning models. The clustering module autonomously identifies topics and groups service requests without requiring service personnel to manually review or sample the data, enabling the system to serve itself.

Inventive Principle:

Principle #25Self-service

2Measurement precision

If manual customization and maintenance of rules is required, then correspondence with predefined themes can be determined, but the process is unduly burdensome to maintain

Engineering Contradiction:

Improvetheme correspondence precisionVSAvoidrule maintenance complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces manual rule-based theme correspondence with automated machine learning topic modeling. The system automatically learns themes from unstructured text data and creates topic clusters without requiring manual customization or maintenance of rules, thereby maintaining precision while eliminating maintenance burdens.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system dynamically adapts to new themes and terminology in service requests through continuous learning from unstructured text data. Unlike static rules that require manual updates, the machine learning model automatically evolves to handle new themes, making the system flexible and maintenance-free.

Inventive Principle:

Principle #15Dynamics

3Productivity

If conventional expert selection relies on expert skills databases, then expert selection can be performed, but significant manual intervention is required

Engineering Contradiction:

Improveexpert selection efficiencyVSAvoidautomation level

Core Design Contradiction:

ProductivityVSExtent of automation

Solution Approach 1:

The patent replaces conventional manual expert selection processes with automated machine learning-based recommendation systems. The system automatically analyzes service requests, identifies relevant topics, and selects appropriate experts without requiring manual intervention, significantly improving productivity and automation levels.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system introduces machine learning topic models as an intermediary between service requests and expert selection. This intermediary automatically processes unstructured text data, identifies themes, and facilitates automated expert matching, eliminating the need for manual intervention in the selection process.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Device complexity

If unstructured text data is ignored, then processing can be simplified, but valuable information is lost

Engineering Contradiction:

Improveprocessing complexityVSAvoidinformation loss

Core Design Contradiction:

Device complexityVSLoss of information

Solution Approach 1:

The patent replaces the approach of ignoring unstructured text data with automated machine learning-based analysis. The system processes and extracts valuable information from unstructured service request summaries through natural language processing and topic modeling, maintaining information integrity while managing processing complexity through automation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10235452B1Expert recommendation leveraging topic clusters derived from unstructured text data

Publication Date: 2019.03.19 EMC IP HLDG CO LLC

AI summary

An apparatus comprises a processing platform configured to implement an expert recommender engine. The expert recommender engine receives information relating to a communication from a user device, and identifies at least one subject matter expert for the communication based on the received information and unstructured text data of a service events database. The expert recommender engine is associated with a clustering module that separates the unstructured text data into topic clusters. The expert recommender engine comprises a collaborative filtering module that receives the topic clusters from the clustering module and utilizes the topic clusters to identify the subject matter expert. The user device is connected with an expert device corresponding to the identified subject matter expert. The expert recommender engine may utilize structured data, social media data and customer satisfaction survey data in combination with the received information and the topic clusters to identify the subject matter expert.