Please use this identifier to cite or link to this item:
https://ah.lib.nccu.edu.tw/handle/140.119/74621
題名: | Multi-dimensional Multi-granularities Data Mining for Discovering Innovative Healthcare Services | 作者: | 姜國輝 Chiang, Johannes K;Chu, Chia-Chi |
貢獻者: | 資管系 | 關鍵詞: | Multidimensional Data Mining; Healthcare Services; Customer relationship Management (CRM); Association Pattern, Granular Computing. | 日期: | Mar-2013 | 上傳時間: | 16-Apr-2015 | 摘要: | Data Mining is getting increasingly important for discovering association patterns for health service innovation and Customer Relationship Management (CRM) etc. Yet, there are deficits of existing data mining techniques. Since most of them perform a plain mining based on predefined schemata through the data warehouse as a whole, a re-scan must be done whenever new attributes are added. Secondly, an association rule may be true on a certain granularity but fail on a smaller one and vise verse. Last but not least, they are usually designed to find either frequent or infrequent rules. After a survey of a category of significant health services, we propose a data mining algorithm alone with a forest data structure to solve aforementioned weaknesses at the same time. At first, we construct a forest structure of concept taxonomies that can be used for representing the knowledge space. On top of it, the data mining is developed as a compound process to find the large-itemsets, to generate, to update and to output association rules that can represent services portfolio. After a set of benchmarks derived to measure the performance of data mining algorithms, we present the performance with respect to efficiency, scalability, information loss, etc. The results show that the proposed approach is better than existing methods with regard to the level of efficiency and effectiveness. | 關聯: | Journal of Data Processing,3(1),31-37 | 資料類型: | article |
Appears in Collections: | 期刊論文 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
243-737-1-PB.pdf | 673.93 kB | Adobe PDF2 | View/Open |
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.