Spark Architecture and Fractional Artificial Bee Colony-Chaotic Fruitfly Ride NN for Big data Classification in Internet of Things
Naeem Th. Yousir
Naeem Th. Yousir*, College of Information Engineering, Al-Nahrain University, Baghdad, Iraq.
Manuscript received on September 22, 2019. | Revised Manuscript received on October 05, 2019. | Manuscript published on October 30, 2019. | PP: 542-555 | Volume-9 Issue-1, October 2019 | Retrieval Number: A9667109119/2019©BEIESP | DOI: 10.35940/ijeat.A9667.109119
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The typical Internet of Things (IoT) device gathers a huge amount of data specifically termed as big data framework, which transfers the collected data from the sensing layer to the information processing layer. Various big data classification methods are adopted in the industrial applications, and smart cities, but accurately classifying the data in the IoT network poses a challenging task in the research community. Therefore, an effective big data classification model using spark-based architecture is proposed in this research. The big data classification is performed at the master node using the proposed Fractional Artificial Bee Colony- Chaotic Fruitfly Rider Optimization Algorithm (FABC-CFFR ide NN). The concept of fictional computing is adopted by the rider optimization algorithm (ROA) to update the position of rider groups based on success rate and the foraging behavior of fruit flies along with the rider parameters is used to enhance to performance of data classification using the proposed CFFR ide NN classifier. Moreover, the proposed Fractional Artificial Bee Colony- Chaotic Fruitfly Rider Optimization Algorithm attained better performance using the metrics, namely accuracy, specificity, and sensitivity with the values of 95.382%, 95.81%, and 98.824% for training percentage without node velocity.
Keywords: Cluster Head (CH), Rider optimization algorithm (ROA), Chaotic Fruitfly algorithm (CFFO), Bhattacharya distance, Big data classification.