Sign in Subscribe
Data mining

New Research on Unstructured Data Clustering Yields Promising Results

Strategic Insights

— 1 min read

Researchers from Jawaharlal Nehru Technological University Kakinada have designed a new clustering-based unstructured data analysis model using a deep learning algorithm. The model, which incorporates a Layer Improved Transformer Network (LITN) and Deep Learning-based Adaptive Clustering (DLAC), has achieved impressive results in terms of clustering performance. The researchers' approach utilizes a Fitness-based Wild Geese Migration with Cuttlefish Algorithm (FWGMCA) to optimize the parameters of the Deep Neural Network (DNN) used in the DLAC. The experimental results show that the designed FWGMCA-DLAC model outperforms traditional clustering approaches, achieving 94.73% Dunn Index, 94.61% Jaccard Index, 89.53% Silhouette, and 94.56% Hopkins values for the second dataset.

Key Takeaways:

  • The researchers have designed a new clustering-based unstructured data analysis model using a deep learning algorithm, which combines a Layer Improved Transformer Network (LITN) and Deep Learning-based Adaptive Clustering (DLAC).
  • The model utilizes a Fitness-based Wild Geese Migration with Cuttlefish Algorithm (FWGMCA) to optimize the parameters of the Deep Neural Network (DNN) used in the DLAC.
  • The experimental results show that the designed FWGMCA-DLAC model outperforms traditional clustering approaches, achieving high Dunn Index, Jaccard Index, Silhouette, and Hopkins values.
  • The model has been tested on two datasets, with the second dataset showing the highest performance.
  • The researchers have confirmed that the implemented FWGMCA-DLAC highly improves the clustering performance than the existing approaches.
  • The model has been peer-reviewed and published in the journal Cluster Computing.

Statistics:

  • 94.73% Dunn Index
  • 94.61% Jaccard Index
  • 89.53% Silhouette
  • 94.56% Hopkins values
  • 2 datasets used in the experimental results
  • The FWGMCA-DLAC model outperforms traditional clustering approaches

Sources:

  • Cluster Computing (2025; 28(13))
  • Jawaharlal Nehru Technological University Kakinada
  • NewsRx LLC

Read more