Please use this identifier to cite or link to this item:

Title: A novel chance model for building innovation diffusion scenario
Authors: Hong, C.-F.;Lin, M.-H.;Yang, Hsiao-Fang;Huang, C.-J.
Contributors: 資管系
Keywords: Chance building;Chance discovery;Innovation diffusion;Rare association rules mining;Weak-Tie;Association rules;Cybernetics;Data storage equipment;Diffusion;Statistical methods;Technology;Videodisks;Innovation
Date: 2010
Issue Date: 2015-05-20 16:23:18 (UTC+8)
Abstract: In order to stay in business, nowadays companies frequently face the problem of choosing new technologies in making new products. To stay in the trend, firms need to sift out useful new technologies from available technologies to gain competitive edges. In this research, we propose a chance model to help creating an innovation diffusion scenario. Three different filtering methods were developed in this model: first, frequent knowledge and outlier were ruled out using upper bound and lower bound of rare chances which were determined by the statistical distribution of skewness. Second, a new event detection method filtered out rare but older event while maximum entropy method filtered out rare but inactive technologies. Third, the diffusion path crossing chasm was discovered and its social influence guaranteeing a successful scenario was estimated. Case study showed the trend of DVD, which followed our model to discover an innovation diffusion scenario; the weak-tie, LIETON-IT, is the diffusion path for diffusing the new technology, H-H DVD-DUAL, into the DVD-DUAL social network. Besides, our model also discovered LIETON-IT owns stronger social influence power to diffuse H-H DVD-DUAL into the social network. The experimental results evidence the usefulness of our model in building innovation diffusion scenario. ©2010 IEEE.
Relation: Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
Data Type: conference
DOI 連結:
Appears in Collections:[資訊管理學系] 會議論文

Files in This Item:

File SizeFormat

All items in 學術集成 are protected by copyright, with all rights reserved.

社群 sharing