Analysis of Internet Customer Complaints Based on Category and Time
碩士 === 輔仁大學 === 資訊工程學系 === 98 === With the development and prevalence of the Internet, the users begin to exchange various types of information about the use of a product over the Internet. At present, the goal of analyzing customer complaints is to build a "database of response strategies"...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2010
|
Online Access: | http://ndltd.ncl.edu.tw/handle/54819604786736551434 |
id |
ndltd-TW-098FJU00392039 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-098FJU003920392015-10-13T18:25:53Z http://ndltd.ncl.edu.tw/handle/54819604786736551434 Analysis of Internet Customer Complaints Based on Category and Time 以分類與時間為基礎之網際網路客訴分析 Kaiming Liu 劉凱銘 碩士 輔仁大學 資訊工程學系 98 With the development and prevalence of the Internet, the users begin to exchange various types of information about the use of a product over the Internet. At present, the goal of analyzing customer complaints is to build a "database of response strategies" so that the best approach can be quickly provided when the customer service center of a company or factory receives a complaint from a customer. However, the traditional method is often unable to provide information such as the level of importance, processing priority, or convenience from a consumer’s point of view to enterprises while facing customer complaints. This paper proposes an Internet customer complaints analysis model that takes the category and time of complaints into consideration. The type of a complaint, the time of its existence, as well as the product stage it belongs to, are all taken into account for the analysis. Based on these data, a weight is computed by the proposed analysis model and according to the weight, Internet customer complaints are sorted to provide a more useful reference for businesses facing these complaints, or for consumers making decisions on goods purchases. Based on the proposed Internet customer complaints analysis model, 172 Internet customer complaint files have been analyzed. The overall result shows that our classification accuracy rate is 93.02%. Furthermore, after cross-checking the weight calculated from categorization and time with the original complaint data and characteristics of the product, the proposed model indeed reflects more effectively the priority of customer complaints, allowing enterprises to voluntarily grasp the causes of the problems much better than analysis methods employed in the past. Hsing Mei 梅興 2010 學位論文 ; thesis 86 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 輔仁大學 === 資訊工程學系 === 98 === With the development and prevalence of the Internet, the users begin to exchange various types of information about the use of a product over the Internet. At present, the goal of analyzing customer complaints is to build a "database of response strategies" so that the best approach can be quickly provided when the customer service center of a company or factory receives a complaint from a customer. However, the traditional method is often unable to provide information such as the level of importance, processing priority, or convenience from a consumer’s point of view to enterprises while facing customer complaints.
This paper proposes an Internet customer complaints analysis model that takes the category and time of complaints into consideration. The type of a complaint, the time of its existence, as well as the product stage it belongs to, are all taken into account for the analysis. Based on these data, a weight is computed by the proposed analysis model and according to the weight, Internet customer complaints are sorted to provide a more useful reference for businesses facing these complaints, or for consumers making decisions on goods purchases.
Based on the proposed Internet customer complaints analysis model, 172 Internet customer complaint files have been analyzed. The overall result shows that our classification accuracy rate is 93.02%. Furthermore, after cross-checking the weight calculated from categorization and time with the original complaint data and characteristics of the product, the proposed model indeed reflects more effectively the priority of customer complaints, allowing enterprises to voluntarily grasp the causes of the problems much better than analysis methods employed in the past.
|
author2 |
Hsing Mei |
author_facet |
Hsing Mei Kaiming Liu 劉凱銘 |
author |
Kaiming Liu 劉凱銘 |
spellingShingle |
Kaiming Liu 劉凱銘 Analysis of Internet Customer Complaints Based on Category and Time |
author_sort |
Kaiming Liu |
title |
Analysis of Internet Customer Complaints Based on Category and Time |
title_short |
Analysis of Internet Customer Complaints Based on Category and Time |
title_full |
Analysis of Internet Customer Complaints Based on Category and Time |
title_fullStr |
Analysis of Internet Customer Complaints Based on Category and Time |
title_full_unstemmed |
Analysis of Internet Customer Complaints Based on Category and Time |
title_sort |
analysis of internet customer complaints based on category and time |
publishDate |
2010 |
url |
http://ndltd.ncl.edu.tw/handle/54819604786736551434 |
work_keys_str_mv |
AT kaimingliu analysisofinternetcustomercomplaintsbasedoncategoryandtime AT liúkǎimíng analysisofinternetcustomercomplaintsbasedoncategoryandtime AT kaimingliu yǐfēnlèiyǔshíjiānwèijīchǔzhīwǎngjìwǎnglùkèsùfēnxī AT liúkǎimíng yǐfēnlèiyǔshíjiānwèijīchǔzhīwǎngjìwǎnglùkèsùfēnxī |
_version_ |
1718033171608502272 |