Feature and Subfeature Selection for Classification Using Correlation Coefficient and Fuzzy Model

2021 ◽  
pp. 1-15
Author(s):  
Hemanta Bhuyan ◽  
Dr. Chinmay Chakraborty ◽  
Subhendu Pani ◽  
Vinayakumar Ravi
Keyword(s):  
Correlation Coefficient ◽  
Fuzzy Model ◽  
Selection For

Relationships between body weight and some follicle and fleece characters in an Australian fine-wool Merino flock

1969 ◽  
Vol 20 (6) ◽  
pp. 1177 ◽  
Author(s):  
T Nay ◽  
RH Hayman
Keyword(s):  
Body Weight ◽  
Correlation Coefficient ◽  
Unit Area ◽  
Additive Effect ◽  
Phenotypic Correlations ◽  
Close Relationship ◽  
Simultaneous Selection ◽  
Selection For ◽  
Fleece Weight ◽  
Highly Correlated

Phenotypic correlations between body weight, follicle characters, and fleece characters have been investigated in a flock of 95 fine-wool non-Peppin Merino wethers, reared in the same locality under the same conditions. The results indicated that three follicle characters, follicle chord, follicle depth, and the index of follicle bending, were independent of body weight. It was found, in agreement with the results of other workers, that body weight was highly correlated with both greasy and clean wool weights (r = 0.53 and 0.51 respectively). It was also found that follicle characters were significantly correlated with most of the fleece characters which contribute to the clean fleece weight per unit area of skin. The correlation coefficient of wool weight per unit area of skin with follicle chord was 0.33, with follicle depth 0.28, and with follicle bending index –0.35. The follicle characters were also correlated with greasy and clean fleece weights. Crimp number per inch was predicted for individual animals by using as criterion the length of the follicle chord. A highly significant correlation coefficient of 0.67 was obtained between predicted and observed crimp number. The close relationship between crimp chord and follicle chord reported in previous work was confirmed. It is suggested that simultaneous selection for body weight and certain follicle characters may have an additive effect on the production of clean wool. It is also suggested that the genetic antagonism between clean wool weight and number of crimps per inch can be explained in anatomical terms.

scholarly journals (231) Correlation of Cluster Number and Tendril Density in Grapevines

HortScience ◽  
2005 ◽  
Vol 40 (4) ◽  
pp. 1008C-1008
Author(s):  
Peter Cousins
Keyword(s):  
Correlation Coefficient ◽  
The United States ◽  
Indirect Selection ◽  
Yield Component ◽  
Cluster Number ◽  
Density Correlation ◽  
Primary Yield ◽  
National Plant Germplasm System ◽  
The Mean ◽  
Selection For

The grapevine shoot has a zone in which leaf-opposed clusters are found at the nodes. Beyond the cluster zone, leaf-opposed tendrils are borne at the nodes in a patterned distribution. Cluster number is a primary yield component and selection programs for increasing yield in grapevine frequently consider cluster number. However, selection for increased cluster number requires direct observation, which is only possible once the vine matures. Clusters and tendrils are developmentally related, so it may be that tendril density (tendrils per node) reflects cluster number. In contrast to cluster number, tendril density can be observed on plants of all ages. The hypothesis that tendril density is related to cluster number was tested here. Cluster numbers and tendril density were assessed on 10 primary shoots each of 180 grapevine (Vitis) accessions. The accessions analyzed are cultivars and wild species collections held in the United States National Plant Germplasm System. The correlation coefficient of the number of clusters and tendril density was calculated using the means of 10 observations per accession. Tendril density was determined by calculating the mean number of tendrils per node in the nodes beyond the cluster zone. Cluster number and tendril density were positively correlated; the correlation coefficient was 0.35. This implies that vines with more tendrils per node also tend to have more clusters. The positive correlation of cluster number and tendril density has implications for grapevine improvement, pointing to the possibility of indirect selection for higher cluster number through selection for higher tendril density. Correlation between juvenile tendril density and mature cluster number is yet to be tested.

IMAGE SHAPE COMPARISON IN MOSAIC DIFFUSION MORPHOLOGIES

2021 ◽  
pp. 3-14
Author(s):  
Yu. V. Vizilter ◽  
O. V. Vygolov ◽  
S. Yu. Zheltov
Keyword(s):  
Correlation Coefficient ◽  
Morphological Analysis ◽  
Fuzzy Model ◽  
Mathematical Form ◽  
Unified Approach ◽  
Shape Comparison ◽  
Mosaic Model ◽  
Image Shape ◽  
The Subject ◽  
Morphological Correlation

In this paper, developing the early proposed unified approach to the representation of morphological models, we show that since morphological models in the attribute representation have the same mathematical form as images, the morphological models themselves can be the subject of morphological analysis and can be directly compared in form with using morphological operators. In this case, the previously introduced formalism of mosaic diffusion morphology is used.In the framework of mosaic diffusion morphology, two alternative descriptions of the projection of the image on the form are considered, which are based on a clear model and a fuzzy model of the form respectively. It is shown that the projection operator in the second case is a one-sorted diffuse operator that makes direct comparision of model to model instead of image to model. In this case, a fuzzy mosaic model appears in this scheme as a projection of a clear mosaic model onto another clear mosaic model. Based on this shape-to-shape projection idea, we propose the new version of Pytiev morphology tools for shape comparison: the morphological shape difference map, the morphological quasi-distance between shapes, as well as the Morphological Shape Correlation Coefficient (MSCC). We show that MSCC from the resource parameters of the reciprocal model has exactly the same formula as the standard effective morphological correlation coefficient proposed earlier based on statistical averaging of projected images.

scholarly journals Presenting a Method for Evaluating the Safety Risk Management Process in Industries Using Fuzzy Logic Approach

2021 ◽  
Author(s):  
Hamidreza Mehri ◽  
Faeze Sepahi-Zoeram ◽  
Ali Karimi ◽  
Farideh Golbabaei
Keyword(s):  
Risk Management ◽  
Fuzzy Logic ◽  
Correlation Coefficient ◽  
Content Validity ◽  
Fuzzy Model ◽  
Face Validity ◽  
Management Process ◽  
Alpha Coefficient ◽  
Safety Risk ◽  
Risk Management Process

Background: An effective process for preventing industrial accidents basically requires a thorough study of the environment, data collection, evaluation, and analysis of this information, determination of corrective action, and its implementation. Risk management provides an integrated framework for this important process. The purpose of this study was to identify the parameters of the risk management process, combine these parameters by fuzzy logic and construct a fuzzy model to obtain the risk management index and finally design a questionnaire with Likert scale to obtain the inputs of this model to evaluate the risk management process. Methods: This descriptive cross-sectional study was conducted in 2018 in Tehran. First, based on library studies and experts' opinions, Jaques non-linear crisis management model was selected, and based on this model, the parameters of the risk management process were extracted. Then, a questionnaire with 22 questions was designed to measure these parameters, the content and face validity of which were evaluated. Also, to evaluate the reliability of the questionnaire, the test-retest method and Cronbach's alpha coefficient were used. Then the parameters were defined as fuzzy numbers, the Fuzzy inference engine was programmed using fuzzy rules, and its validity was evaluated. Results: The fuzzy model has three stages, in each of which sixteen rules are used. In this fuzzy model, the defuzzification step was performed by four methods with the same results. The designed questionnaire contains twenty-two questions, the content validity ratio (CVR) for this questionnaire is 0.89, and the content validity index (CVI) for all questions was above 0.79. Cronbach's alpha coefficient for this questionnaire was 0.713. Face validity was determined quantitatively by calculating the impact score (more than 1.5). Using intraclass correlation coefficient and Pearson correlation coefficient, the existence of reliability between test times (test-retest) was confirmed, so that their values were 0.84 and 0.88%, respectively. Conclusion: The proposed fuzzy model has a high validity giving a correct evaluation of the risk management process and expressing the final result in the form of an index between zero and one hundred. The risk management process evaluation questionnaire has good validity and reliability with the interpretation that the item has good face validity and is understandable, simple, and fluent for the sample group. Using this tool, industry managers can evaluate the safety risk management process, making them able to identify the strengths and weaknesses of this process, and finally take steps to eliminate the defects and improve this process continuously.

Sign in / Sign up

Export Citation Format

Share Document