| Author: | Wang, Gang |
| Title: | Data stream mining : techniques and applications |
| Degree: | M.Sc. |
| Year: | 2007 |
| Subject: | Hong Kong Polytechnic University -- Dissertations. Data mining. Computer security. Streaming technology (Telecommunications) |
| Department: | Department of Computing |
| Pages: | xiii, 264 leaves : ill. ; 30 cm. |
| Language: | English |
| Abstract: | Currently, many and many organizations have realized the advantage that data mining technologies brings on their business. In many domains, data mining provide the competitive ability of intelligence for today's business activities. Data mining technologies have created many successful cases, have achieved lots and lots of interests on the research and industry in the recent years. But in order to fast make a decision and response to extra-world, we have to process streaming data in real time to discover the interesting knowledge or find anomaly varying of data streams. Different from the data sets to which traditional data mining technologies face, data streams are transient data that pass through the system continuously and in huge volumes. There are many applications that require handling data in the form of a stream, such as sensor network data in a modern car, RFID network based on modern logistic flow and SCM (Supply Chain Management), network traffic flow, time-series data, stock exchange data, telecommunications, Web click streams, weather or environment monitoring, and so on. In the latter environment, usually there is only limited resource such as finite memory, lower speed CPU, distribution sensor nodes etc. Therefore, our research should focus on the systematic investigation of stream data mining principles and algorithms, and research the effective, efficient, and scalable methods for mining the dynamics of data streams, this thesis will develop and implement some new and existing algorithms to discover changes, trends and evolution characteristics in data streams, construct clusters and classification models from data streams, and explore frequent patterns and similarities among data streams. The methods developed by this thesis will be applied to Power System Fault Recording System, SCM with RFID or other distribution sensor network, financial data trend prediction, and other applications. Especially, on Supply Chain Management (SCM), information system has been broadly employed in recent years. The theories and technologies of SCM trend to the perfection and maturation. But yet there is much open problem need to solve. In this thesis, we will depict the related problem with a division of retail SCM. In retail SCM, there are yet different degree inefficiencies over upstream or downstream direction. Upstream inefficiencies result in high out-of-stock situation, low return rate as well as long lead times. Downstream inefficiencies affect demand forecast accuracy, low on-shelf availability and thus loss of revenue despite the fact that products are available on site. Furthermore, information-sharing restrictions between trading partners reduce the accuracy of demand forecast and the scheduling of the replenishment process. Thus, how find a model to solve the above these issues? In this thesis, we will propose a model "RFID + Events Detection + Data streams management + Data streams mining" to address the above these issues through creating a SCEM System. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b21288768.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 17.77 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/3309