Integrating Naïve Bayesian Classification and Internet of Thing into Semantic-Based Cloud Computing
碩士 === 國立虎尾科技大學 === 資訊工程系碩士班 === 105 === This study proposed an integrating Naïve Bayesian classification semantic cloud computing framework (INBCSCCF) to construct a system-based framework by using the Internet of Things and sensing technology. Hadoop-based semantic web technology was employed as t...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2017
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/25q5v2 |
| Summary: | 碩士 === 國立虎尾科技大學 === 資訊工程系碩士班 === 105 === This study proposed an integrating Naïve Bayesian classification semantic cloud computing framework (INBCSCCF) to construct a system-based framework by using the Internet of Things and sensing technology. Hadoop-based semantic web technology was employed as the core of the backend system. Subsequently, data were collected through the IoT framework and the Naïve Bayesian classifier was used to analyze the data. Hadoop-based MapReduce distributed computing was adopted to solve the performance problem of computing large volumes of data. A culture sharing cloud platform (CSCP), which was developed using cultural and creative activities as the basis, was used to verify the feasibility of this study’s INBCSCCF. The development of the CSCP was focused on integrating an information processing system of the exhibition site with a brainwave sensor and mobile phone to develop the IoT framework required for the system to achieve communication between objects in a wireless network environment. The sensor data were analyzed using the Naïve Bayesian classifier to produce data regarding users at the exhibition. Semantic web was adopted to enhance the interactive relationship between different types of information. Machine learning technology was applied to the sensing data generated from the CSCP to determine the characteristics of user preference. The results verified that the proposed INBCSCCF can solve the problems of deriving from the sensing data collected using the system.
|
|---|