Evaluation of Calinski-Harabasz Criterion as Fitness Measure for Genetic Algorithm Based Segmentation of Cervical Cell Nuclei

Cengizler, Caglar; Kerem-Un, M

doi:10.9734/bjmcs/2017/33729

Cited by 15 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The algorithm was able to detect close relationships between questions, with small between-cluster variance indicating similar clusters. The Calinski-Harabasz score [55] was 124.89, which is relatively low, indicating that the clusters were not well separated, and may not be distinct. There was a variation within the Calinski-Harabasz score when the assigned number of neighbors (n-neighbors) was between 2 and 19.…”

Section: Semantic Analysis and Manual Comparison To Standardize The C...mentioning

confidence: 96%

DataPLAN: a web-based data management plan generator for the plant sciences

Zhou,

Beier,

Brilhaus

et al. 2023

Preprint

View full text Add to dashboard Cite

Research Data Management (RDM) is a system for the effective handling of research data that enables scientists to structure their research questions and ensure best practices throughout the data lifecycle, from acquisition, computation and annotation to data publication and re-use. Data management plans (DMPs) are documents that formally set out the RDM of a project and are required by many funding bodies. DMPs help to organize and structure RDM strategies, thus promoting data findability, accessibility, interoperability and reusability (FAIR). Although DMPs incorporate methods and standards that can be reused by different research projects, the standardization of DMP content is not as evident as the standardization of RDM practices and data/metadata. To address this issue in the plant sciences, we developed DataPLAN — a tool that combines a questionnaire with pre-written standardized responses. We wrapped the questionnaire in a serverless single-page web application that can then generate standardized responses from DMP templates. The current templates cater to plant research grant proposals for Horizon 2020, Horizon Europe and the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG). In the future the range of templates will be extended to accommodate other funding schemes, thereby enabling more users to generate their own templates. The DataPLAN web application is open-source and does not require an internet connection. By utilizing DataPLAN, the workload associated with creating, updating, and adhering to DMPs is significantly reduced.

show abstract

Section: Semantic Analysis and Manual Comparison To Standardize The C...mentioning

confidence: 96%

DataPLAN: a web-based data management plan generator for the plant sciences

Zhou,

Beier,

Brilhaus

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Other works applied different CVIs, e.g., Xiao et al [ 38 ] proposed a hierarchical K-means algorithm that incorporates the Davies–Bouldin index as a metric, enabling the efficient identification of the number of clusters while minimizing computation time and costs. Caglar et al [ 39 ] showed the effectiveness of the Calinski–Harabasz index as a robust cluster validation measure; it surpassed the performance of the Jaccard index and F-score in the clustering evaluation of cervical cells. Some relevant works on human posture recognition have also employed the aforementioned CVI for both supervised and unsupervised learning techniques [ 40 , 41 , 42 , 43 , 44 , 45 ].…”

Section: State Of the Artmentioning

confidence: 99%

Comparative Analysis of the Clustering Quality in Self-Organizing Maps for Human Posture Classification

Ekemeyong Awong,

Zielinska

2023

Sensors

View full text Add to dashboard Cite

The objective of this article is to develop a methodology for selecting the appropriate number of clusters to group and identify human postures using neural networks with unsupervised self-organizing maps. Although unsupervised clustering algorithms have proven effective in recognizing human postures, many works are limited to testing which data are correctly or incorrectly recognized. They often neglect the task of selecting the appropriate number of groups (where the number of clusters corresponds to the number of output neurons, i.e., the number of postures) using clustering quality assessments. The use of quality scores to determine the number of clusters frees the expert to make subjective decisions about the number of postures, enabling the use of unsupervised learning. Due to high dimensionality and data variability, expert decisions (referred to as data labeling) can be difficult and time-consuming. In our case, there is no manual labeling step. We introduce a new clustering quality score: the discriminant score (DS). We describe the process of selecting the most suitable number of postures using human activity records captured by RGB-D cameras. Comparative studies on the usefulness of popular clustering quality scores—such as the silhouette coefficient, Dunn index, Calinski–Harabasz index, Davies–Bouldin index, and DS—for posture classification tasks are presented, along with graphical illustrations of the results produced by DS. The findings show that DS offers good quality in posture recognition, effectively following postural transitions and similarities.

show abstract

“…This choice is also based on our analysis using two concrete statistical scores and allows us to better tune our unsupervised learning process. Two well known statistical clustering scores the (a) silhouette (S) score [50] and (b) Calinski-Harabasz (CH) score [51] were used for choosing the optimization parameter k for all data sets. These have been chosen based on their high intrinsic differences.…”

Section: Clustering and Optimizationmentioning

confidence: 99%

2D MoS₂ nanopores: ionic current blockade height for clustering DNA events

et al. 2019

View full text Add to dashboard Cite

2D nanopores can be used to electrophoretically drive DNA molecules, which can in turn be identified through measurable electronic current blockades. In this work, we use experimental data from molybdenum disulfide nanopores threading DNA nucleotides and propose a methodological approach to interpret DNA events. Specifically, the experimental ionic traces are used to train an unsupervised machine learning model for identifying distinct molecular events through the 2D nanopore. For the first time, we propose a clustering of experimental 2D nanopore data based on the ionic current blockade height and unrelated to the traditional dwell time for each DNA event. Within this approach, the blockade level information is implicitly included in the feature space analysis and does not need to be treated explicitly. We could show the higher efficiency of the blockade height over the traditional dwell time also in coping with sparse nanopore data sets. Our approach allows for a deep insight into characteristic molecular features in 2D nanopores and provides a feedback mechanism to tune these materials and interpret the measured signals. It has, thus, a high impact on the efficiency of 2D nanopore-based DNA sequencers.

show abstract

Evaluation of Calinski-Harabasz Criterion as Fitness Measure for Genetic Algorithm Based Segmentation of Cervical Cell Nuclei

Cited by 15 publications

References 10 publications

DataPLAN: a web-based data management plan generator for the plant sciences

DataPLAN: a web-based data management plan generator for the plant sciences

Comparative Analysis of the Clustering Quality in Self-Organizing Maps for Human Posture Classification

2D MoS₂ nanopores: ionic current blockade height for clustering DNA events

Contact Info

Product

Resources

About

Evaluation of Calinski-Harabasz Criterion as Fitness Measure for Genetic Algorithm Based Segmentation of Cervical Cell Nuclei

Cited by 15 publications

References 10 publications

DataPLAN: a web-based data management plan generator for the plant sciences

DataPLAN: a web-based data management plan generator for the plant sciences

Comparative Analysis of the Clustering Quality in Self-Organizing Maps for Human Posture Classification

2D MoS2 nanopores: ionic current blockade height for clustering DNA events

Contact Info

Product

Resources

About

2D MoS₂ nanopores: ionic current blockade height for clustering DNA events