Browse Prior Art Database

Method and System for Analyzing User Behavior and Identifying High-value Key Words in Self-Service based Helpdesk Solution

IP.com Disclosure Number: IPCOM000198187D
Publication Date: 2010-Jul-29
Document File: 5 page(s) / 128K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method and system for analyzing user activities on self-help applications in service desk and recommending high-value key words to users in a novel manner, intended to reduce service desk support cost and to increase end users productivity and satisfaction.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 5

Method and System for Analyzing User Behavior and Identifying High -

Solution

Disclosed is a method and system for analyzing user activities on self-help applications in service desk and recommending high-value key words to users in a novel manner, intended to reduce service desk support cost and to increase end users productivity and satisfaction. We're using Hidden Marcov Model (HMM) to estimate user's visit type by category (i.e. success, no solution, helpdesk, and others), and extracting and identifying hot and valuable key words from successful visits.

Self-Service based helpdesk solution didn't get as many adoptions as assumed, because it's hard for end users to figure out a proper key word and perform the precise searching while using the website. End users have voiced frustration over some self-service solutions, primarily with the search functionality. The known solution is that while end-user is experiencing some technical problems, they search most of solution documents with some of key words, and explores different parts of solution to find out what they really need to solve their issues. If they don't find what they want then they will continue trying out other key words. If they still have not found after trying many times, they will open a ticket or start a chat with helpdesk to seek their help until this issue is resolved by them. The drawback is that when end-users have some of the urgent problems, sometimes they don't immediately find the right solution to solve their problems. Even worse, they spent lots of time on finding solution without any relevant results, and this may seriously affect their working efficiency and productivity, as well as increase ticket number of creating
(i.e. IT support cost), etc.

To address this top challenge of those users with the self-service based helpdesk solutions,

major components Visit Estimator, Key Word Extractor, and Key Word Recommender. This system evaluates user visit types (i.e. success, no solution, helpdesk, etc) and extract the high value key words from the successful visits and recommend them to other users that may experience similar problems, as depicted in Figure 1 below.

-value Key Words in Self

value Key Words in Self -

-Service based Helpdesk

Service based Helpdesk

we design a system including three

1

Page 2 of 5

2

[This page contains 1 picture or other non-text object]

Page 3 of 5

Figure 1 - Overall scenario and flow for analyzing user visit types and recommending key words

The core idea of this disclosure is summarized as follows.

1) Using HMM(Hidden Markov Models) to predict the success of user visits and refine KB and searching, it uses HMM for evaluating whether th...