STATE OF ART OF MACHINE LEARNING FOR STREAMING DATA

  • Unique Paper ID: 162067
  • Volume: 9
  • Issue: 7
  • PageNo: 935-946
  • Abstract:
  • The era of big data has given rise to an unprecedented influx of streaming data, generated continuously and in real-time from various sources such as social media, sensors, and IoT devices. Traditional machine learning algorithms designed for static datasets face significant challenges when applied to streaming data due to its dynamic and evolving nature. This research paper explores the paradigm of machine learning for streaming data, focusing on adaptive techniques that can handle the continuous and high-velocity flow of information.Various adaptive machine learning techniques are reviewed, including online learning algorithms, incremental learning, and concept drift detection methods. The paper provides a comprehensive overview of how these methods enable models to evolve and adapt in real-time, ensuring their relevance in dynamic environments. Additionally, the paper explores emerging technologies and frameworks that facilitate the implementation of adaptive machine learning for streaming data, such as Apache Flink, Apache Storm, and online learning libraries. The challenges associated with deploying these techniques in real-world scenarios, such as resource constraints and scalability, are also addressed. In conclusion, this research paper contributes to the ongoing discourse on machine learning for streaming data by presenting a comprehensive overview of adaptive techniques and their applications.
email to a friend

Cite This Article

  • ISSN: 2349-6002
  • Volume: 9
  • Issue: 7
  • PageNo: 935-946

STATE OF ART OF MACHINE LEARNING FOR STREAMING DATA

Related Articles