A new affinity matrix weighted k-nearest neighbors graph to improve spectral clustering accuracy

Article Computer Science, Information Systems

Multilevel Graph-Based Decision Making in Big Scholarly Data: An Approach to Identify Expert Reviewer, Finding Quality Impact Factor, Ranking Journals and Researchers

Muhammad Mazhar Ullah Rathore et al.

Summary: Scholarly Big Data generated by digital libraries consists of conference papers, journal documents, books, and thesis, covering scholarly information from researcher's and publisher's perspectives. The relationships among Big Scholarly Data are crucial for solving researcher and journal concerns, with the graph being the best approach for efficient processing. Proposed mechanisms include selecting suitable reviewers, designing journal ranking based on modified page ranking algorithm, calculating researchers ranking, and processing big graphs using parallel processing mechanism over the real-time analysis approach.

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING (2021)

添加到收藏夹

Article Computer Science, Hardware & Architecture

Optimized clustering-based discovery framework on Internet of Things

Monika Bharti et al.

Summary: With the advancement of technology, the emergence of Internet of Things has provided a method for automated interactions between users and their environment, but interactions among humans and devices remain challenging. The study proposed an OCDF-IoT framework, utilizing optimized clustering for intelligent resource discovery and efficient management. Results showed that the framework performed effectively in resource indexing and selection, demonstrating high efficiency and stability.

JOURNAL OF SUPERCOMPUTING (2021)

添加到收藏夹

Article Computer Science, Hardware & Architecture

Ensemble-based clustering of large probabilistic graphs using neighborhood and distance metric learning

Malihe Danesh et al.

Summary: This paper introduces a novel method based on ensemble clustering for large probabilistic graphs, which involves generating ensemble clusters and utilizing a probabilistic co-association matrix to integrate base clustering results. Two improvements are applied in the process to enhance the accuracy of probability estimation and learn crucial features from ensemble clusters. Evaluation on real-world datasets demonstrates the impressive performance of clustering large probabilistic graphs.

JOURNAL OF SUPERCOMPUTING (2021)

添加到收藏夹

Article Chemistry, Multidisciplinary

ORC Optimal Design through Clusterization for Waste Heat Recovery in Anaerobic Digestion Plants

Guido Francesco Frate et al.

Summary: The study proposed a design methodology for ORC based on the boundary conditions expected during the operation of anaerobic digestion plants, and compared four different designs, showing that the design obtained from the most populated cluster generated 10% more energy than that from a system designed based on year average conditions.

APPLIED SCIENCES-BASEL (2021)

添加到收藏夹

Article Chemistry, Multidisciplinary

A Novel Load Forecasting Approach Based on Smart Meter Data Using Advance Preprocessing and Hybrid Deep Learning

Fatih Unal et al.

Summary: Short-term load forecasting models are crucial for distribution companies to make effective decisions, especially when forecasting load profiles of many end-users at the customer-level which faces challenges such as high variability and uncertainty. The novel hybrid deep learning approach for energy consumption prediction outperforms traditional prediction models, showing higher accuracy and robustness.

APPLIED SCIENCES-BASEL (2021)

添加到收藏夹

Article Computer Science, Information Systems