Indonesian Journal of Electrical Engineering and Computer Science Vol. 25, No. 2, February 2022, pp. 1159~1166 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v25.i2.pp1159-1166  1159 Journal homepage: http://ijeecs.iaescore.com Different analytical frameworks and bigdata model for internet of things Ayushi Chahal, Preeti Gulia, Nasib Singh Gill Department of Computer Science and Applications, Maharshi Dayanand University, Rohtak, Haryana, India Article Info ABSTRACT Article history: Received Mar 26, 2021 Revised Dec 15, 2021 Accepted Dec 22, 2021 Sensor devices used in internet of things (IoT) enabled environment produce large amount of data. This data plays a major role in bigdata landscape. In recent years, correlation, and implementation of bigdata and IoT is being extrapolated. Nowadays, predictive analytics is gaining attention of many researchers for big IoT data analytics. This paper summarizes different sort of IoT analytical platforms which consist in-built features for further use in machine learning, MATLAB, and data security. It emphasizes on different machine learning algorithms that plays important role in big IoT data analytics. Besides different analytical frameworks, this paper highlights the proposed model for bigdata in IoT domain and elaborates different forms of data analytical methods. Proposed model comprises different phases i.e., data storing, data cleaning, data analytics, and data visualization. These phases cover the basic characteristics of bigdata V’s model and most important phase is data analytics or big IoT analytics. This model is implemented using an IoT dataset and results are presented in graphical and tabular form using different machine learning techniques. This study enhances researchers’ knowledge about various IoT analytical platforms and usability of these platforms in their respective problem domains. Keywords: Bigdata Big IoT data analytics Frameworks IoT Machine learning This is an open access article under the CC BY-SA license. Corresponding Author: Ayushi Chahal Department of Computer Science and Applications, Maharshi Dayanand University Rohtak, Haryana, India Email: ayushichahal@gmail.com 1. INTRODUCTION In today’s era, internet of things (IoT) and bigdata are potential areas of research. Heterogeneous data is generated by the different sensor’s devices in the IoT environment. Bigdata analytics is used to search, mine, and analyze IoT data. It can also be used to handle structured, semi-structured and unstructured data [1] and helps to convert this data into some understandable form for the analysis. There are multiple techniques that can be used for bigdata analytics such as classification, clustering, association rules, and prediction. Bigdata is characterized by 10 V’s: Volume, Velocity, Variety, Value, Veracity, Validity, Variability, Viscosity, Virality and Visualization [2] shown in Figure 1. Gartner defines bigdata concept that helps in decision making, optimizing the processes, discover patterns insightfully. He gave a characteristics model for bigdata which defines three V’s that is volume, velocity and variety of data. Gartner research has made estimation that by 2022 most of the data generation and its analysis will be done by machines rather than humans. So, it is need of an hour to have a model which can handle Big IoT data efficiently for prediction using Machine learning techniques.