Using Dynamic Neural Network Input Item To Predict Customer 's Behavior
碩士 === 淡江大學 === 資訊工程學系 === 88 === With the rapid development of computer and Internet technology, there is a great change for aspects of business, entertainment and education. Besides the development of technology, people start to think about how to apply these advanced technologies. In t...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2000
|
Online Access: | http://ndltd.ncl.edu.tw/handle/14989676925508896022 |
id |
ndltd-TW-088TKU00392037 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-088TKU003920372016-01-29T04:19:18Z http://ndltd.ncl.edu.tw/handle/14989676925508896022 Using Dynamic Neural Network Input Item To Predict Customer 's Behavior 使用動態類神經網路輸入項預測客戶行為 Wei-Herng Jeng 鄭維恆 碩士 淡江大學 資訊工程學系 88 With the rapid development of computer and Internet technology, there is a great change for aspects of business, entertainment and education. Besides the development of technology, people start to think about how to apply these advanced technologies. In this paper, we propose a dynamic neural network input item model to predict customer’s behavior on bond fund. The accuracy of our model is about 90 percent for testing data. Moreover, the model has proved to be valuable by sales-people in real investing operations that would be in 74 percent accuracy. The advantage of the proposed model is the original net system does not have to be reconstructed completely whenever some new entities added into or deleted in the process. Our proposed model constructed with environmental objects and individual entities. Therefore, we can use some partial variations of entities to predict the future trends of other entities, or change environmental objects to observe the individual behavior of entities. In fact, it is helpful in decision-making by enhanced understanding of potential factors on entity’s behavior Ding-An Chiang 蔣定安 2000 學位論文 ; thesis 63 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 淡江大學 === 資訊工程學系 === 88 === With the rapid development of computer and Internet technology, there is a great change for aspects of business, entertainment and education. Besides the development of technology, people start to think about how to apply these advanced technologies. In this paper, we propose a dynamic neural network input item model to predict customer’s behavior on bond fund. The accuracy of our model is about 90 percent for testing data. Moreover, the model has proved to be valuable by sales-people in real investing operations that would be in 74 percent accuracy. The advantage of the proposed model is the original net system does not have to be reconstructed completely whenever some new entities added into or deleted in the process. Our proposed model constructed with environmental objects and individual entities. Therefore, we can use some partial variations of entities to predict the future trends of other entities, or change environmental objects to observe the individual behavior of entities. In fact, it is helpful in decision-making by enhanced understanding of potential factors on entity’s behavior
|
author2 |
Ding-An Chiang |
author_facet |
Ding-An Chiang Wei-Herng Jeng 鄭維恆 |
author |
Wei-Herng Jeng 鄭維恆 |
spellingShingle |
Wei-Herng Jeng 鄭維恆 Using Dynamic Neural Network Input Item To Predict Customer 's Behavior |
author_sort |
Wei-Herng Jeng |
title |
Using Dynamic Neural Network Input Item To Predict Customer 's Behavior |
title_short |
Using Dynamic Neural Network Input Item To Predict Customer 's Behavior |
title_full |
Using Dynamic Neural Network Input Item To Predict Customer 's Behavior |
title_fullStr |
Using Dynamic Neural Network Input Item To Predict Customer 's Behavior |
title_full_unstemmed |
Using Dynamic Neural Network Input Item To Predict Customer 's Behavior |
title_sort |
using dynamic neural network input item to predict customer 's behavior |
publishDate |
2000 |
url |
http://ndltd.ncl.edu.tw/handle/14989676925508896022 |
work_keys_str_mv |
AT weiherngjeng usingdynamicneuralnetworkinputitemtopredictcustomersbehavior AT zhèngwéihéng usingdynamicneuralnetworkinputitemtopredictcustomersbehavior AT weiherngjeng shǐyòngdòngtàilèishénjīngwǎnglùshūrùxiàngyùcèkèhùxíngwèi AT zhèngwéihéng shǐyòngdòngtàilèishénjīngwǎnglùshūrùxiàngyùcèkèhùxíngwèi |
_version_ |
1718169019668758528 |