Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Author: search.filters.author.Shetty D, D.Subject: search.filters.subject.Clustering

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Novel schemes for energy-efficient IoT
    (Springer Verlag service@springer.de, 2019) Venkateshwarlu, K.; Shetty D, D.
    Internet of things (IoT) is a global infrastructure for the information society which enables advanced services by interconnecting physical and virtual things based on existing and evolving inter-operable information and communication technologies. Developing green IoT is a difficult task because IoT has more devices and has complex structure, so most of the current schemes for deploying nodes in wireless sensor networks (WSNs) cannot be applied directly in IoT. In this paper, we propose a scheme which gives an energy-efficient IoT. Here, we propose two schemes for framework structure of a network, and then we propose clustering algorithms and routing algorithms for network formation which is based on minimum spanning tree. After numerous simulations, we show that these schemes result in minimal energy consumption and enhance the network lifetime. Thus, the proposed schemes are more energy-efficient compared to a typical WSN deployment scheme; hence, these schemes are applicable to the green IoT deployment. We show that in the proposed schemes, the nodes are alive for more number of rounds as compared to the existing algorithms. © 2019, Springer Nature Singapore Pte Ltd.
  • No Thumbnail Available
    Item
    An Approach for Efficient Graph Mining from Big Data Using Spark
    (Springer Science and Business Media Deutschland GmbH, 2023) Gupta, R.K.; Shetty D, D.; Chakraborty, S.
    Huge amount of data is generated and accumulated over the last decade, and therefore, the use of data mining techniques is required to extract usable information from these massive data sets. Gaining important connections between data helps in getting useful insights. Depiction of relationships between the data using graphical approach is observed to be a helpful method. It provides an effective technique for demonstrating the working in a variety of situations, including biological networks, social networks, Web networks, and so on. Clustering techniques used in graph mining can be helpful for accumulating significant information. In this paper, an approach for graph mining from big data in Spark (AGMBS) is proposed on the basis of label propagation. The suggested technique enhances the efficiency of the conventional label propagation algorithm by making it more resilient. In addition to this, AGMBS employs a sparse matrix as its primary data structure, resulting in quicker performance. Thereafter, GraphX is used for managing the processing of the graphical data. The experiments were conducted on two graph data sets from the real world, and it is observed that the suggested AGMBS gives faster results as compared to the best available clustering algorithms. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.